UNIVERSITY Of TORONTO
• JAN 1 S 1927
DEPT. OF
MATALLURGiCAL ENGiNEERiNG
r"
DEPARTMENT OF METALLURRIOAL ENGIMF-ERING
library NiimHrrs •
Return thi ■ L -
Cuf*i)vx.irdt
Sh :f»
is are to Uj sirred f;r in Uiii iua.i
^j.,
vui ioT a iu.i^.-r
THE IRON AND STEEL
MAGAZINE
SUCCESSOR TO THE METALLOGRAPHIST
A MONTHLY PUBLICATION DEVOTED TO THE
IRON AND STEEL INDUSTRY
EDITED BY
ALBERT SAUVEUR
VOLUME X
JULY TO DECEMBER, 1905 ^i^/^l 1
ROTCH BUILDING
CAMBRIDGE, MASSACHUSETTS
// 1
JULY, 1Q05
PAGE
The Application of Dry-Air Blast James Gayley i
Note on the Failure of an Iron Plate through
Fatigue " Sidney A. Houghton . . 11
Chemical Changes in the Open-Hearth Furnace . W. M. Carr 17
Copper Alloys L. Guillet 21
Metallography Applied to Foundry AVork . . . Albert Sauveur .... 29
A New Hardening Furnace t^t^
The Economic Value of Cast Iron IT'. H. Pretty 37
The Constitution of Iron-Carbon Alloys .... E. Heyn 42
Abstracts 53
Metallurgical Notes and Comments 68
Review of the Iron and Steel Market 80
Statistics 84
Recent Publications 86
Patents 93
AUGUST, 1905
Paul Louis Toussaint Heroult Frontispiece
Some Causes of Failure of Rails in Service . . . Robert 'fob 97
A Comparison of vStandard Methods for Testing
Cast Iron Dr. Richard Moldenke .. 107
Recent Developments of the Bertrand-Thiel Pro-
cess in the Manufacture of Steel . . John H. Darby and George Hatton 112
The Continuous Steel Process in Fixed Furnaces . S. Surzycki 118
Copper Alloys L. Guillet 124
Vanadium and Vanadium Steel 134
Metal Mixers for Pipe Foundries J. B. Nau 141
Melting Steel with Cast Iron R. P. Cunninghayn ... 145
Rail Sections as Engineering vStructures .... P. H . Dudley 149
Hard Cast Iron : A Theory of One of its Causes . Henry Souther 154
Etching of High Carljon Steel E. H. Saniter 156
Abstracts 157
Metallurgical Notes and Comments 165
Review 01- the Iron and Steel Market 180
Statistics 183
Recent Publications 186
Patents 191
ii
Coiitoits HI
SEPTEMBER, 1905
PAGE
Rossitor Worthington Raymond Frontispiece
Descriptive Metallurgy of Iron and Steel . . . Suniiid (iroves 193
The Manufacture and Characteristics of Wrought
Iron James /'. Roe 199
The Thermit Process in American Practice . . . Ernest Stittz 212
The Application of Dry-Air Blast to the Manu-
factvire of Iron 7". IT. Robinson .... 224
The Influence of Titanium on Pig Iron and Steel . Pierre Delville 230
Protection of Iron and Steel Structures .... I.ouis H. Barker .... 234
High-Speed Tool Steels /-./'. Breckenridge ... 237
Abstr.vcts 247
Met.\llurgical Notes and Comments 250
Review of the Iron and Steel Market 274
Statistics 278
Recent Publications 282
Patents 287
OCTOBER, 1905
Sir Lowthian Bell Frontispiece
Electric Steel F. IF. Harhord .... 289
The Galbraith Electric Iron and Steel Furnace 294
The Melting Points of Slags and Other Members
of the Series Si02-Al.,0:t-CaO Clifford Richardson ... 297
Descriptive Metallurgy of Iron and Steel . . Samuel Groves 300
Metallography Applied to Foundry Practice . . Albert Sauveur .... 309
Special Steels L. Guillet 314
Hot Cracks in Steel Castings \rthur Simonson ... 321
Cost of Producing Steel Castings by the Open-
Hearth Process and the Small Converter . . /.. Unckenbolt 324
Xew Open-Hearth Steel Process P. Ackers ....... 327
The Cleaning of Blast- Furnace Gas Axel Sahlin ^33
Abstracts 345
Metallurgical Xotes and Comments 354
Review of the Iron and Steel Market 371
Statistics 375
Recent Publications 377
Patents 384
NOVEMBER, 1905
Robert Forrester Mushet Frontispiece
Overheated Steel Arthur Windsor Richards and John Edward Stead 385
iv Contents
PAGE
Xew Gin Process for the Electrical Manufactiire
of Steel Gustave Gin 404
Steel as an Igneous Rock 40S
Metallography Applied to Foundry Work . . . Albert Sauvcur .... 413
Crystalline and Amorphous States of Metals 41Q
Abstracts 426
Metallurgical Notes and Comments 443
Review of the Iron and Steel Market 463
Statistics • 467
Recent Publications 473
Patents 479
DECEMBER, 1905
Sir Henry Bessemer Frontispiece
The Genesis of the Bessemer Process 481
Dry Air for Blast Furnaces 502
Iron Resources of the World R. Anspach 511
Shall We Substitute Iron for Steel ? 519
Great Britain's Iron Industry T. Good 523
Open-Hearth Furnace Comparison.s A. D. Williams, Jr. . . 533
Abstracts 538
Metallurgical Notes and Comments 546
Review of the Iron and Steel Market 561
Statistics 565
Recent Publications 571
Patents 575
See pages 6§ and 66
/.
The Iron and Steel Magazine
Je veux au mond publier
d'une plume de fer sur un papier d'acier."
Vol. X
July, 1905
No. I
THE APPLICATION OF DRY-AIR BLAST (Supplementary Paper)
By JAMES GAYLEY, New York
T T is to be regretted that the data respecting the use of dry-air
blast, which were presented to the Institute at its meeting
in the United States, in Octo-
ber, should have been re-
stricted to the period from
August 25 to September 9
inclusive, and from Septem-
ber 17 to 30 inclusive. In
order to present the paper
at all it was found necessary
to limit the record of opera-
tions to the period above
stated. In the discussion of
the paper it would appear
from the conclusions of some
participating therein, that
they have been placed at
some disadvantage in considering the economies obtained through
the use of dry-air blast, by reason of the data covering such a
short period. It is the purpose of this communication to present
in detail the record of operations of the Isabella Furnaces from
November, 1904, to March, 1905, inclusive, as shown by the fur-
nace records.
The month of October is not included, as the furnace using
dry-air blast was stopped several times for repairs, and while
The Iron and Steel Magazine
Table I. — November.
Grains of Moisture.
Teinpe
rature.
Gas Analysis.
Date.
In Atmosphere.
In Dry
Blast.
Atmosphere.
Dry Blast.
CO.
CO2.
Day.
Night.
Day.
Night.
Day.
Night.
Day.
Night.
Nov. 1
2-53
2-74
1 15
105
51
47
20
17
25-8
12-6
. 2
2-82
2-75
110
102
50
50
18
16
25-6
122
. 3
2-64
2-59
105
101
47
45
16
16
23-6
14 0
, 4
218
2-44
106
107
52
50
16
19
23 0
15 0
. 5
2-58
1-98
113
101
54
44
17
19
. 6
1-69
1-79
0-94
ro6
45
44
15
19
. 7
1-86
2-35
105
1-04
43
41
17
20
22-8 .
14 0
. 8
2-42
2-43
104
104
45
44
20
20
, 9
2-30
2-39
109
102
46
42
19
19
. 10
2-52
2-49
1-07
105
44
45
19
20
22-8
14 0
. 11
1-69
1-68
100
104
44
39
18
20
. 12
1-50
173
107
li)2
42
47
20
22
, 13
1-88
179
104
1-00
43
44
21
22
. 14
1-56
1-59
104
0-95
44
41
20
21
23-8
13-8
. 15
1-89
1-91
107
106
45
45
20
22
. 16
2 01
1-91
111
102
49
39
20
21
20 -6
13 0
. 17
1-74
1-85
101
1-05
41
39
18
22
, 18
2-04
2-26
113
109
48
48
21
22
23-4
14 0
. 19
2-24
2-35
110
109
48
50
21
22
. 20
2-74
3 03
116
104
59
69
22
20
. 21
1-96
1-85
0-98
0-96
52
39
18
20
23-2
16 0
. 22
2 06
2-02
105
103
43
40
20
20
. 23
2-21
2-18
103
101
44
51
20
21
24 -0
12 0
. 24
2-43
1-81
107
1-00
48
44
21
21
. 25
1-62
1-32
0-94
0-99
43
44
22
21
24-8
12 0
, 26
1-49
1-43
098
101
39
35
20
21
, 27
101
0-90
0-93
0 81
33
29
19
17
, 28
0-88
102
0-81
0-83
31
37
18
18
22-6
15 0
. 29
2-33
2-30
107
103
56
49
21
20
, 30
105
100
0-81
0-82
37
33
17
18
23 0
15 0
Average
1-99
1-99
103
1 01
46
43
19
20
23-5
13-8
The record of operations is as follows :-
Average
Daily
Product.
Average
Coke Con-
sumption.
Blowing-
Engines.
Revolutions
per Minute.
Average
Temperature.
Hot Blast.
No. 1 Furnace (Dry Air) .
No. 3 Furnace (Natural Air)
Tons.
447
386
Lbs.
1816
2279
96
111
Deg
854
750
TJie Application of Dry- Air Blast
the record was a very good one considering these interruptions,
yet it was not a continuous one.
The record for No. i Furnace using dry blast, and the com-
parison with No. 3 Furnace using normal blast for the month of
November, 1904, is set forth in Table I.
Table II.-
—December.
Date.
Grains of Moisture.
Temperature.
Gas Analysis.
In Atmosphere.
In Dry Blast.
Atmoj
sphere.
Dry
Blast.
CO.
CO2.
Day.
Night.
Day.
Night.
Day.
Night.
Day.
Night.
Dec. 1
113
1-37
0-98
107
35
36
20
21
23-4
13-6
.. 2
1-41
1-56
1-08
0-97
38
37
20
20
.. 3
1-56
1-46
102
0-96
35
34
20
20
.. 4
1-26
1-30
0-99
0-97
33
29
19
19
23-0
h'o
.. 5
1-34
1-39
0-97
0-98
31
36
18
20
23 0
14 0
.. 6
1-28
1-50
100
1-00
35
36
20
20
25-0
120
.. 7
1-70
1-67
1-10
0-99
39
40
20
20
.. 8
VM
1-20
1-00
0-95
41
34
20
19
22-6
14 -8
.. 9
112
111
0-93
0-85
32
29
18
18
24-8
14-4
.. 10
0-91
0-55
0-76
0-52
25
15
16
11
24 0
14 0
.. 11
0-55
0-80
0-43
0-44
15
23
7
10
.. 12
1-26
1-37
0-85
0-94
30
30
17
19
24-4
12-8
.. 13
1-04
0-83
0 93
0-64
29
23
18
14
23 0
13-6
.. 14
0-73
0-64
0-67
0-53
22
13
12
9
24 0
14 0
.. 15
074
0-78
0-53
0'56
22
20
8
10
.. 16
0-79
0-99
0-72
070
20
21
10
11
23 -0
15 -0
.. 17
139
1-65
0-79
1-02
29
34
14
20
.. 18
1-37
1-72
1-00
106
32
32
19
20
., 19
1-81
1-35
111
0-96
35
31
21
18
223
15-0
.. 20
1-29
1-31
102
1-04
31
34
18
19
22-8
13-4
.. 21
1-34
1-27
107
1-00
31
30
19
18
22-4
13-6
.. 22
1-23
1-91
110
1-08
35
49
19
20
22-6
144
.. 23
2-44
3-50
1-35
1-54
53
53
23
27
., 24
2 15
1-71
118
0-94
40
35
21
18
23 -1
13-2
.. 25
2 19
2-31
0-96
103
37
38
17
19
.. 26
2-37
3 06
111
1-25
41
46
19
22
.. 27
3-91
2-23
1-41
111
58
48
24
21
., 28
0-68
0-72
0-74
0-60
25
21
15
13
247
13-3
.. 29
0-82
0-94
0-67
0-69
25
29
13
13
24-2
130
..30
104
1-26
0-82
0-91
38
43
15
16
24-8
136
.. 31
155
2-22
0-96
111
47
51
19
22
Average
1
1-41
1-47
0-94
0-92
34
33
17
18
23-5
18-8
Note. — All temperatures are Fahrenheit, and tons are 2240 lbs,
in vols.
Gas analyses expressed
By a reference to my former paper it will be noted that the
month of November is the beginning, in this part of the country,
of the winter period, when the atmosphere decreases rapidly in
humidity. The months of October and April being the transition
months in the autumn and the spring respectively, furnaces
4 The Iron and Steel Magazine
using natural air approach more closely, from November on, to
the conditions obtained by the use of dry-air blast. The data
given in Table I, with reference to grains of moisture and tem-
perature, represent the average for the day, of observations
taken hourly during the period indicated.
The temperature of the dry blast is from observations taken
at the top of the refrigerator chamber, but this temperature is
increased in the passage of the air from the refrigerator chamber
to the blowing engines.
In Table II is shown the record of operations for the month
of December, 1904, and in this record it is interesting to note
the decrease in moisture as compared with November. In No-
vember the average moisture was 1.99 grains, while in December
it was 1.45 grains, the maximum variation, however, being from
0-55 to 3.91 grains.
There is also a slight reduction in the grains of moisture in
the dry air.
The air conduit pipe from the refrigerator chamber to blow-
ing engine room was constructed to connect with four blowing-
engines, and as only three engines were used in No. i Furnace,
it was decided to connect the fourth engine with the dry-air
conduit and apply it to No. 3 Furnace, thus making on No. 3
two engines with natural air and one engine w4th dry air. The
comparison of the work of No. i w^th No. 3 Furnace, and the
effect of one third of the engine revolutions supplying dry air
to No. 3 Furnace, is shown as follows:
No I Furnace :
Dry Blast
No. 3 Furnace :
Dec. 1-22 (Normal Blast)
Dec. 23-31 (^ Dry Blast)
Although one third of the engine revolutions supplied dry
air, yet the weight of dry air was slightly in excess of one third
the total quantity, by reason of the air being denser.
The result shown by the use of such a small quantity of dry
air is remarkable, and is greater than was experienced when
first applying the dry air to No. i Furnace in August. The
Averac?e
Daily
Product
Average
Coke Con-
sumption
Blov/ing Engines
Revolutions
per Minute
Average
Temperature
Hot Blast
Tons
Lbs.
Deg.
455
1,823
96
877
400
2,309
(III)
(785;
461
2,140
llie Application of Dry-Air Blast
application of one third of the engine revolutions with dry air
increased the weight of air delivered to the furnace, and at the
same time the average temperature of hot blast increased 20°.
The furnace at once began to drive more rapidly, and the burden
was also increased. There was no deterioration in the grade of
Table III
— January
Date.
Grains of Moisture.
Temperature.
Gas Analysis.
In Atmosphere.
In Dry
Blast.
Atmosphere.
Dry Blast.
CO.
COj.
Day.
Night.
Day.
Night.
Day.
Nighi.
Day.
Night.
Jan. 1
2-61
2-75
119
114
53
51
22
20
.. 2
3-31
2-19
1-34
0-82
52
41
21
15
,
. 3
0-98
0-52
0-50
0-45
27
18
9
9
' 24-8
13 0
. 4
0-70
0-78
0-68
0-57
20
22
10
10
24-3
13-2
. 5
109
1-34
073
078
29
33
13
14
23 0
14-2
. 6
1-85
1-96
0-94
0-80
33
35
16
16
24 0
13 0
. 7
1-49
119
0-81
0-68
32
29
15
14
23-4
13-0
. 8
1-00
0-89
071
070
26
23
13
12
. 9
0-97
1-72
0-69
0-84
25
31.
12
15
.;
, 10
0-61
0-90
0-58
0-48
21
23-
10
7
. 11
1-65
2-89
0-66
0-81
31
45
9
12
. 12
3 05
1-51
100
0-66
48
37
15
11
. 13
100
0-86
0-68
0-48
31
27
10
8
...
. 14
0-53
0-65
0-49
0-49
17
20
7
8
. 15
0-77
0-63
0-59
0-45
24
19
9
7
,16
0-85
0-95
0-61
0-53
23
28
10
10
. 17
108
1-23
0-65
0-64
31
34
12
12
. 18
1-31
1-33
075
0-69
36
41
12
13
22-8
14-2
. 19
2 05
1-96
0-86
0-86
42
42
15
16
. 20
1-67
1-65
078
0-64
42
36
13
11
22-6
14'0
. 21
1-87
2-15
076
075
40
39
12
13
,
. 22
1-43
0-79
077
0-48
35
25
13
9
. 23
0-85
1-20
0-56
0-57
22
26
9
11
23'-8
13'8
. 24
1-44
0-97
0-64
051
29
23
12
9
. 25
0-67
0-42
0-45
0-35
19
12
7
4
23'0
12-8
. 26
0-47
0-69
0-38
0-42
15
17
5
6
22-2
13 0
. 27
0 94
118
0-46
0-53
24
30
7
7
23-8
13-5
. 28
0-69
0-38
0-50
0-27
23
9
8
2
22-6
14-8
. 29
0-47
0-53
0-28
0-36
10
18
-1
3
. 30
072
072
0-41
0-37
18
15
3
3
23"4
is'e
. 31
0-99
1-22
0-50
0-55
22
28
5
8
23-3
167
Average
1 26
1-23
0-67
0-60
29
28
11
10
23-3
13-5
metal produced, in fact it was slightly better, as the silicon was
higher and sulphur lower than in the preceding part of the month.
In Table III is set forth the record for January, 1905, whe in
is found a further reduction in the humidity of the atmosphere.
On January 10, the dry blast was changed from No. i to
No. 3 Furnace. Both furnaces were making the same grade of
6 The Iron and Steel Magazine
iron for use in the basic open-hearth process. In the table
below no account is taken of the period from the nth to 14th
inclusive, as this period was occupied in adjusting the dry
blast and burden on both furnaces. The record of the burden
on each furnace shows as follows:
Weight of Coke Weight of Ore
in Charge in Charge
No. I Furnace : Lbs. Lbs.
Jan. i-io (Dry Blast) 10,200 24,000
Jan. 15-31 (Normal Blast) 10,200 20,200
'No. 3 Furnace:
Jan. i-io (Normal Blast *) (Extra coke
in charge) • 10,20c -:o,2oo
Jan. 15-31 (Dry Blast) . . • ■• :,••..; 10,200 23,600
During the period from January i to 10, when No.
3 Furnace was on normal blast, there was charged with the regu-
lar burden a small quantity of extra coke. This was taken off,
wiien the furnace was changed to dry blast, which would make
the burden with dry blast for No. 3 Furnace correspond with
No. I Furnace with dry blast. .The result on each furnace,
before and after the change in blast had been made, is as follows:
Average Average Blowing Average
Daily Prod- Coke Engines Temperature
uct Consump- Revolutions Hot Blast
tion per Minute
No. I Furnace : Tons Lbs, Deg.
Jan. I-IO (Dry Blast) . . 428 1,825 96 869
Jan. 15-31 (Normal Blast) 414 2,340 iii 771
No. 3 Furnace:
Jan. I-IO (Normal Blast) 410 2,351 11 1 7^6
Jan. 15-31 (Dry Blast) . 432 1,811 96 80-?
The ore mixture on No. 3 Furnace gave a yield in iron i per
cent greater than the mixture on No. i Furnace. The purpose
in changing the dry blast from No. i to No. 3 Furnace was to
determine the economy on another furnace at a time which
represented in that locality nearly the minimum of humidity in
the_ atmosphere, the extreme variations being from 0.38 to 3.31
grains, with a monthly average of 1.25 grains of moisture per
cubic foot of air. The response to the application of dry-air
blast was prompt and efficient in result, and clearly demonstrated
that, even at periods when the humidity of the atmosphere was
relatively low, substantial economy in fuel could be obtained.
* January i-io only.
TJic Application of Dry- Air Blast 7
As already pointed out, an important advantage obtains by
keeping the moisture in the air more uniform. Although the
atmosphere is much more humid in the summer than in the winter
months, yet in the latter the percentage of variation is much
greater. A comparison of the average humidity of the several
months of the year is misleading as to the effect of such humidity
on the operations of a furnace, since such average results do not
take into account the wide fluctuations from d^y to- day, and
even in the same day. During the past winter iii the Pittsburg
district, and generally throughout the northern states, there has
prevailed a protracted cold season, and since the beginning of
our observations there has been no winter season showing a lower
average of humidity than the one just ended ; therefore any dem-
onstration under these conditions of the efficiency of dry-air blast
in comparison with a furnace using normal blast is made at a
time when the furnace operated with normal blast is at its highest
efficiency.
It has been thought by some that the use of dry-air blast
might be dispensed with in the winter months, when the content
of moisture in the atmosphere is very low. There could scarcely
be a month more favorable to the study of the effect of dry air
than that presented in the month of February, 1905, where the
average of moisture is 1.19 grains for the daytime and 1.17 grains
for the night, with a maximum variation of 0.30 to 2.57 grains of
moisture per cubic foot of air. In Table IV is shown the record
for that month, and it will be noted that during several days the
average moisture content of the atmosphere is lower than the
average for the month in the dry blast.
As already pointed out the benefits derivable from the use
of dry air can be directed in the main to increase of production,
or to decrease in coke consumption, or to both purposes. In
the month of February, by reason of the dry atmosphere and
the large volume of air on No. i Furnace, the operations at No.
3 Furnace were directed principally toward economy in coke.
The output on No. 3 is less than No. i Furnace, as the former
was stopped several times during the month on account of break-
downs of slag machine and breakouts of iron at the hearth.
Notwithstanding the low Content of moisture in the normal blast,
representing conditions rarely obtained in that district, the fur-
nace supplied with dry blast made, considering the stops, prac-
The Iron and Steel Magazine
Table IV. — February.
Date.
Grains of Moisture.
Temperature.
Gas Analysis.
In Atmosphere.
In Dry
Blast.
Atmosphere.
Dry Blast.
CO.
CO2.
Day.
Night.
Day.
Night.
Day.
Night.
Day.
Night.
Feb. 1
1-33
0-73
0-71
0-47
30
21
10
7
23-6
134
.. 2
0-39
0-30
0-35
018
13
7
3
-1
M 3
0-48
0-39
0-31
0-22
9
9
1
1
23-8
i3-4
., 4
0-58
0-49
0-39
0-28
16
14
3
4
.. 5
0-73
1-64
0-41
0-63
20
31
6
11
.. 6
1-66
0-79
0-71
0-49
34
23
13
7
24 0
13 0
.. 7
0-69
0-67
0-51
0-35
23
16
7
3
.. 8
0-79
2-27
0-42
0-80
26
39
4
14
24-6
140
.. 9
2-57
1-67
0-95
0-78
43
37
16
13
23-4
14 0
.. 10
102
0-61
0-83
0-46
30
20
13
17
.. 11
0-60
104
0-56
0-52
18
30
6
9
,, 12
2-25
2 01
0-87
0-81
37
37
12
13
.. 13
0-72
0-40
0-64
0-28
20
9
9
2
24-6
13-4
M 14
0-40
0-71
0-37
0-35
8
21
0
5
., 15
0-45
0-32
0-48
024
16
:9
3
2
.. 16
0-53
0 73
0-35
0-33
15
27
2
4
23 6
12-2
., 17
104
0-84
0-61
0-48
33
27
9
6
240
14-2
., 18
0-73
0-68
0-55
0-39
25
24
6
4
. . •
.. 19
0-87
115
0 51
0-60
27
32
5
9
., 20
1-98
2-06
0-86
0-97
39
41
12
13
.
.. 21
1-56
211
0-71
0-88
38
37
14
10
25-6
12-4
.. 22
2-37
2 01
1 02
0-86
40
38
12
10
.. 23
1-86
1-69
111
0-74
28
36
13
9
24 0
14 0
.. 24
1-70
165
0-88
0-70
37
32
9
8
22 0
15 0
.. 25
1-77
2 10
0-91
0-88
34
40
10
10
. .•
.. 26
1-48
0-91
0-90
0-43
38
28
13
-5
.. 27
0-97
1-34
0-70
0-53
29
35
8
10
.. 28
1-74
1-40
0-62
0-69
40
37
8
7
25-6
12 2
Average
119
117
0-65
0-54
28
27
8
'
23-9
13 4
The record of operations is as follows : —
Average
Daily
Product.
Average
Coke
Consump-
tion.
Blowing-
Engines.
Revolutions
per Minute.
Average
Temperature.
Hot Blast.
No. 1 Furnace (Normal Blast)
No. 3 Furnace (Dry Blast)
Tons.
424
412
Lbs.
2248
1815
Ill
96
Deg.
800
784
The Application of Pry- Air Blast g
tically as much iron — with a consumption of coke 433 pounds
less per ton of iron — as the furnace supplied with normal blast.
As the summer season approaches, the product of the furnace
using normal blast will steadily decrease and the fuel increase,
w^hile the work of the furnace supplied with dry air will continue
practically uniform. It is doubtful if a better illustration can
be had than that offered by the month of February in demonstra-
ting the value of maintaining the blast practically uniform as to
dryness.
As will be seen by Table V the advent of the spring months
makes a material increase in the moisture. The operations in
March were seriously interrupted by reason of high water in the
Allegheny River, which flooded the works, and caused a shut-
down of several days. On account of this shut-down extra
charges of fuel were added to both furnaces. On starting the
plant. No. 3 Furnace, supplied with dry air, responded more
quickly and reached the normal grade of iron in one day, while
it required three da3^s to obtain like result on No. i Furnace
The record for the month is set forth in Table V.
The product on No. 3 Furnace is a little less than on No. i
Furnace, but No. 3 was banked nearly one day longer than No. i,
which would more than make up the difference.
In the discussion on my paper presented to the Institute in
October, 1904, 1 note a comparison of the Isabella with the Edgar
Thomson furnaces, without due regard being given to the differ-
ent conditions existing at these plants. During the period
covered in this communication the Edgar Thomson furnaces
worked with a mixture of ores yielding 55.5 per cent of iron,
while the Isabella mixture yielded but 51.5 per cent of iron, and,
in addition, the stoves at the Edgar Thomson gave a tempera-
ture of from 200° to 300° higher than at the Isabella. The
results achieved by the use of dry blast have been from furnaces
that might properly be designated as old furnaces; they have
been banked several times, which invariably has a deteriorat-
ing effect, and the coke used was an inferior grade from the Con-
nellsville region, and used altogether on furnaces making basic
iron.
The data presented in this communication are the furnace
records entire for the points covered, and in considering them
it is important to bear in mind that comparisons made with the
lO
The Iron and Steel Magazine
Table V. — March.
Grains of Moisture.
Tempe
rature.
Gas Analysis.
Date.
In Atmosphere.
In Dry
Blast.
Atmosphere.
Dry Blast.
CO.
CO2.
Day.
Night.
Day.
Night.
Day. Night.
Day.
Night.
Mar. 1
1-77
0-78
0-90
0 52
37
27
13
5
.. 2
0-78
100
0-55
0-44
29
30
4
7
. 3
1-81
2-06
0-74
0-59
37
42
9
10
. 4
1-98
116
0-91
0-42
44
31
13
8
...
. 5
1-60
1-69
0-74
0-69
39
40
11
12
. 6
124
1-44
0-75
0-62
38
38
10
10
23-5
14 0
. 7
2-28
3 29
0-90
104
41
47
13
14
, 8
273
1-61
107
0-68
44
33
15
9
. 9
1-73
219
0-68
0-83
35
41
9
12
. 10
1 94
1-20
0-88
0-60
43
32
14
10
...
. 11
1-37
1-52
075
075
39
36
12
11
. 12
133
109
074
0 54
40
32
11
9
...
. 13
100
116
0-71
0-64
35
34
10
12
. 14
1-22
1 32
0-81
0-57
36
33
12
10
. 15
1-64
1-85
0 85
0-84
39
38
10
13
..
. 16
2-47
2-74
111
112
49
52
15
18
. 17
2-64
2-74
1-21
1 12
54
53
18
18
23-9
13-5
. 18
2-85
3-84
1-31
1-38
64
65
21
24
. 19
4-80
408
1-65
1-54
e@
61
24
23
. 20)
■ ^4
. 22j
, 23
Furnace bankec
i on acco
unt of hi{
jh water.
2-61
2 95
0-81
1-08
52
56
la
15
, 24
3-32
3 08
1-32
1-15
58
51
21
16
. 25
2-69
2-73
1-29
109
60
54-
18
16
, 26
319
2-93
1-20
111
58
58
17
16
...
. 27
2-34
2 58
1-26
1-19
64
64
18
17
. 28
2-92
4 04
1-29
1-42
66
63
18
23
. 29
3 08
316
1-34
1-27
68
68
21
23
. 30
3-30
218
1-42
0-88
59
49
22
16
. 31
2-23 j 2-39
1-21
101
57
55
17
15
Av
erage
2 25
2-25
101
0-89
48 I 45
14
14
237
137
The record of operations for the month are as follows : —
Average
Daily
Product.
; No. 1 Furnace —
Normal Blast
No. 3 Furnace —
Dry Blast
Tons.
411
405
Average
Coke
Consump-
tion.
Lbs.
2274
1837
Blowing-
Engines.
Revolutions
per Minute.
Ill
96
Average
Temperature.
Hot Blast.
Deg.
850
784
Xote oil the Faihirc of an Iron Plate' through ''Fatigue " ii
dry blast are under atmospheric conditions when furnaces oper-
ated with normal blast are doing their best work.
The writer takes this opportunity to state that in the dia-
grams in the former paper the record of each day is averaged
with the preceding days.
NOTE ON THE FAILURE OF AN IRON PLATE THROUGH
" FATIGUE " *
By SIDNEY A. HOUGHTON
Assoc. M. Inst. C. E. (London)
\ LTHOUGH the consideration of the failure of metals by what
'^^ is commonly called " fatigue " has engaged the attention
of many experimenters, yet
comparatively little has been
written on the causes which
ultimately produce fracture.
Mr. Thomas Andrews in a
series of articles published
in " Engineering " in 1897-
1898 appears to believe that
failures of this description in
steel are produced by weak-
ening the joints of the crys-
tals accelerated or induced
by cracks arising from the
minute flaws which exist in
all commercial steels, but
this theory is one which has
not' met with general acceptance. More recently Prof.
J. A. Ewing and Mr. J. C. W. Humfrey have given in a
very able paper^f the results of repeated alternations of
stress on the structure of Swedish iron, and they express
the opinion that fracture is due to constant slipping in the
crystals, which ultimately produces cracks in their cleavage
planes. With these two exceptions little or nothing seems to
* Iron and Steel Institute, May, 1905, meeting.
t " Philosophical Transactions," Vol. CC.
12
The Iron and Steel Magazine
have been written as to the causes of fracture under fatigue
stresses, and although it might be thought that with the advent
of microscopical examination it would be a simple matter to
ascertain their effect, yet, with mild steel, at any rate, consider-
able difficulty is experienced owing to the great purity and
ductility of the ferrite crystals, which do not readily show the
effect of small strains. Possibly it may be due to this cause that
the literature on this important subject is so scanty.
Fracture
occurred here
Centre of
Vi/inOirig drum
Fig. I. Sectional Elevation through Back Ring of Barrel of Boiler
In the case about to be described, although the plate was
of iron, its quality differed very materially from that used by
Professor Ewing, and for that reason, and as it occurred in
actual practice, a consideration of it may be of some value.
The plate which failed was in the back ring of the barrel of
a portable boiler of the locomotive type, which formed part of a
steam plowing engine. As a rule barrel plates are not sub-
jected to fatigue stresses of importance, but in this instance the
Note on the Failure of an Iron Plate through ''Fatigue " 13
step or bracket which carried the spindle of the winding drum
was riveted to it (Fig. i), and consequently when plowing was
being done the plate was subjected to severe panting stresses.
These would be in addition to the tensile stress due to the pres-
ure of steam, which would theoretically amount to 3.67 tons per
square inch between the rivet holes, and 2.17 in the solid plate
when new; but owing to wasting, these figures would have in-
creased to 5.71 and 3.38 when fracture occurred. It must also
be noted that the iron was thus stressed when at a temperature
of about 350° F., when its ductility would be less than when cold.
Moreover, as one end of the step was riveted close to a longitu-
dinal lap joint, the fatigue stresses were to a great extent localized,
and failure actually occurred through a crack forming between
the rivet holes. Unfortunately this crack began from inside the
outer lap of the plate, and consequently was not discovered, the
result being that the boiler exploded with great violence and
instantaneously killed a man who was near.
The boiler was about twenty years old, but as plowing is
only done during part of the year the actual time it was at work
may be considered as about six years. The barrel plates w^ere
originally about 7-16 inch thick, but had wasted externally to
9-32 inch near the joint which gave way, conseqtiently the
stresses at this part must have gradually increased in severity.
These plates were made of B. B. iron, the actual analysis being:
carbon, trace; manganese, trace; sulphur, 0.023; phosphorus,
0.310; silicon, 0.180 per cent, from which it will be seen that the
phosphorus is high and the silicon more than usual.
A tensile test cut lengthways of the plate near the fracture
gave 23.8 tons per square inch with only 2 per cent elongation in
i\ inches, the last result being to some extent influenced by a
small flaw near the middle. The fracture was coarse crystalline.
The metal was distinctly hard, and although inferior considered
as iron, its very want of ductility rendered it specially suitable
for studying the effects of fatigue. Several sections were cut
from the plate in three planes, and the structure, which was fairly
similar in all the longitudinal sections, showed, as might be ex-
pected, large ferrite crystals with a considerable quantity of slag
flaws, the slag itself frequently showing a duplex structure.
Many of the crystals exhibited very clearly that wavy eutectic
appearance which seems to be characteristic of commercial
14 The Iron and Steel Magazine
wrought iron containing phosphorus, and to which attention
has already been directed.*
The rivet holes had been punched, and notwithstanding this
was done twenty years ago, the effect on the structure in distort-
ing the neighboring crystals was apparently as clear as if the plate
had just come from the machine. Although failure took place
between the rivet holes, numerous small cracks parallel to the
line of fracture had formed on the inside of the plate in the
vicinity, and it was one of these which, as previously remarked,
reduced the elongation in the tensile test. A fairly typical exam-
ple is shown in Fig. 2, which is a photomicrograph of the plate
about ^ inch from the line of fracture, the top of the photograph
being the inside surface of the plate. The crack begins at the
surface, where it is fairly straight, but it is probable that this part
is influenced by pitting action. A small portion of the crack
appears to the left, but the main part is near the center of the
photograph. Although the different parts appear disconnected,
it must be remembered that only a plane surface is shown, and
it is probable that the gaps are due to these cr3^stals possessing
greater ductility or more suitable orientation to resist fracture,
and that the cracks are really connected in other parts of less
resistance.
The results of the action of the stresses on the structure of
the metal may be summed up as follows :
Formation of slip lines, indicating slipping of the crystals.
Loosening of the joints between the crystals.
Loosening of the particles of slag.
The observance of slip bands in metal is, it is believed, due
to Professor Ewing and Mr. Rosenhain, and they particularly
call attention to their appearance in metal subjected to numerous
reversals of stress. Their remarks are fully borne out by the metal
under consideration, many very fine examples of these lines being
visible in sections near the fracture (Fig. 2). It should, how-
ever, be observed that none were produced in the test piece when
broken in the machine, and this is a somewhat remarkable fact.
In addition to these lines many crystals showed signs of being
strained in various directions, and some even appeared to be
breaking up into small ones. An example of one of these is
* " The Internal Structure of Iron and Steel," by the author. " Pro-
ceedings " of the Institute of Marine Engineers, 1902, Vol. XIV.
Note on the Failure of an Iron Plate through ''Fatigue " 15
shown in Fig. 2, where also may be seen some crystals with wavy
marks.
The looseness at the crystalline joints is only slightly evi-
dent, and would be somewhat difficult to detect without com-
FiG. 2. Longitudinal Section of B. B. Iron Boiler Plate,
showing crack produced by "fatigue." Magnified 53
diameters. Etched with picric acid.
parison with unstrained metal of the same description. The
slag flaws do not seem to have extended appreciably, and the
conclusions arrived at by Mr. Andrews are certainly not sup-
ported by the results of the examination of this plate, which,
1 6 The Iron and Steel Magazine
however, are closely in accordance with the views enunciated by
Professor Ewing; that is, that the real cause of failure is the
fracture of the crystals along cleavage planes. The cracks, how-
ever, appear to start from small surface flaws which would natur-
ally tend to concentrate the stresses, and it would seem,
therefore, important to keep the surface of material subjected to
these stresses as clean and smooth as possible. The cracks seem,
perhaps, to follow the junction lines of the cr^^stals to a rather
greater extent than when metal is broken in a machine, and as
might be expected the slag flaws influence their direction in a
marked manner. Some years ago Mr. Stead called attention to
the fact that probably these flaws helped to increase the resist-
ance of wrought iron to cross-breaking, and this is supported
by the course of the cracks in this metal. Briefly, the effect is
to avoid the concentration of stresses in a single plane, and the
resistance to fractures of this description is, therefore, consider-
ably increased.
As observed, the plate had wasted considerably in the
vicinity of the joint, and it appears probable that this wasting
was to some extent due to the straining of the crystals. For
when slipping occurs minute clean surfaces of iron would be
exposed in close conjunction with the surface oxide, and con-
sequently a rapid electrolytic action would arise. This theory
may also explain the reason of grooving in boilers which occurs
where the plates are locally stressed. It is true that Professor
Bose has shown that when two similar pieces of metal are placed
in an electrolyte and one is strained, an electric current passes,
but it is suggested that the cause of this is due to the exposure of
a fresh surface of metal and not to any undiscovered means for
producing electric potential.
To sum up, the effect of fatigue stresses on this plate has
been to form cracks commencing, as a rule, from irregularities
on the inner surface, which cracks were due to weakness in the
cleavage planes of the crystals from continual slipping, and to a
less degree to some loss of adhesion between the crystals. Some
of the crystals appear to have been broken up, and as regards
the slag flaws these seem to have a restraining effect on the prog-
ress of the cracks.
It may be added that these remarks are not intended to have
a general application, but are written in the hope that they may
y
/
Chemical Chaui^cs in tJic Open-Hearth Furnace 17
lead to further and more comprehensive investigations being
undertaken in this very important subject.
CHEMICAL CHANGES IN THE OPEN-HEARTH FURNACE *
By W. M. CARR
'THHE accompanying diagrams may be of interest in setting
forth the changes of composition occurring in a bath of
molten metal treated under normal conditions of regular open-
hearth practice when producing steel to enter into ordinary cast-
ings. In both cases under investigation the initial chemical
composition of the metals going into each charge was carefully
calculated from numerous analyses of stock on record in the
laboratories of the American vSteel Foundries. The acid heat
was a regular one turned .out at the Chicago works. The basic
heat was the regular product of the Granite City (Illinois) Works.
The investigations were conducted in the early part of 1903
under the writer's direction as metallurgist of their Western
district.
Of course the acid heat was composed of low phosphorus
pig iron and basic steel scrap (billets, plate, defective basic steel
castings and shop scrap). The final analysis of the heat was
made from a sample taken when about half of the contents of the
ladle had gone into castings. This, of course, gave a represen-
tative sample of the entire finished heat. Analyses made at
other times to determine variations of constituents between the
first and last portions of a normal ladleful of steel showed that
for regular practice and purposes a sample taken in the middle
of a pour was decidedly representative of the whole.
All samples taken during the investigation of this acid heat
were regarded as an average at the several times. In following
the manganese line it will be noticed that there is an increase of
that element shortly after the metals are melted. This can be
explained as the result of an addition of a moderate quantity of
ferro-manganese during that period, the aim being to wash the
bath. At the time of the investigation such a plan was regular
practice in all heats at these works. The gradual increase in
* Received in February, 1905.
20 The Iron and Steel Magazine
sulphur would be the result of two causes: First, in view of fuel
oil as a source of heat it was supposed that there would be some
absorption of sulphur from the flame while the metals were
exposed to its action before the final liquefaction and disappear-
ance below the slag and away from its direct influence. Second,
the oxidizing action of the flame would mean some loss in metallic
iron, and in the absence of any active basic absorbent there would
be a corresponding increase of sulphur in the molten metal.
The changes in phosphorus would be subject to the same con-
ditions as the sulphur in the second stage (metals melted) plus
the amounts carried in by the deoxidizing agents, i. e., ferro-
manganese and ferro-silicon.
In regard to the basic heat the conditions are, if anything,
more interesting than in an acid heat, because of greater changes
in composition following the purifying action of CaO furnished
by a liberal charge of limestone. The pig iron of the charge con-
sisted of two kinds of basic pig, one containing about i per cent
phosphorus and the other about 2 per cent. The scrap was
steel rails, defective steel castings and shop scrap. Therefore the
initial phosphorus was not very high in the total charge of metals.
Fuel oil was the source of heat in this case also. It cannot be
said that the samples taken at regular intervals after the com-
pletion of the charge going into furnace and taken immediately
upon the partial melting of the metals were representative ones.
It was the practice to put in last a certain portion of the pig iron,
and owing to its relative low melting point, the samples of metal
taken during the first stage of the investigation (metals melting)
would be those of molten pig iron more or less refined by what-
ever slag may have formed during the first step. The composi-
tion of the mass of metals of course would vary, depending upon
whether one might reach, in taking a sample, a quantity of either
molten pig iron or molten steel scrap. But after complete fusion
and under the layer of rapidly forming slag it was quite safe to
assume, in the gradual progression of the second stage (metals
melted), that the samples would be a fair average of the bath.
The changes in sulphur and phosphorus give plain evidence of
slag action. It is noteworthy that the sulphur decreased rapidly
near the end of the heat. This was thought to be due to the in-
fluence of manganese carried in by the ferro-manganese in the
final working of the heat and the preliminary deoxidizing of the
Copper Alloys 21;
bath before tapping. The change or increment of phosphorus
was due to the influence of siUcon carried in by ferro-silicon used
also as a preliminary deoxidizing agent before tapping, as silicon
would release phosphorus from the slag. The complete deoxidiz-
ing was done in the ladle while tapping with the assistance of
ferro-manganese, ferro-silicon and carborundum. The weights
of carbon shown to be carried in in the final additions were:
furnished by calculations of the available carbon in the several
deoxidizing agents. The same applied to the manganese and
silicon and also refers to the acid heat in the same particulars.
The final analysis was determined on a sample taken about the
middle of the pour while going into castings. This was regular
practice and only represented the steel at that time. In basic
steel for castings it is known that the composition varies in
regard to phosphorus, it increasing towards the end, the cause
being in the main due to a reabsorption of that element by the
metal immediately below the slag. The amount absorbed varies,
greatly and depends upon many conditions. Therefore the final
analysis of the basic heat was fairly representative of the whole,,
with the possible exception of the content of phosphorus.
COPPER ALLOYS*
SPECIAL BRASSES AND QUENCHING OF BRONZE
By L. GUILLET
Translated from the French for The Iron and Steel Magazine
{Continued from page jog, Vol. IX)
Tin Brass — Theoretical Study. Tin brasses have recently
been much used in naval construction. No general study of
this group ot ternary alloys has yet been conducted, and in the
present investigation it has been confined to the influence of tin
upon ordinary brass containing 60 per cent copper and 40 per
cent zinc.
It results from our microscopical study that tin plays the
same part as zinc, but with a much greater activity, i per cent
of tin being as effective as 4 per cent of zinc.
Among the accompanying photomicrographs (Figs. 16 to-
*" Revue de Metallurgie," February, 1905.
22
The Iron and Steel Magazine
19) we shall call attention to those representing an alloy with
54 per cent copper and 6 per cent tin and an alloy with 60 per
cent copper and 10 per cent tin. The latter alloy has the same
Fig. 16. Cu, 60%; Zn, 39-5%; Sn,
0.5%. Magnified 200 diam.
Fig. 17. Cu, 60%; Zn, 39%; Sn, 1%.
Magnified 200 diam.
Fig. 18. Cu, 54%; Zn, 40%; Sn, 6%. Fig. 19. Cu, 60%; Zn, 30%; Sn, 10%.
Magnified 200 diam. Magnified 200 diam.
Structure as the first one although it contains more copper and
more tin. Their structure recalls that of a brass containing
55 per cent zinc.
Copper Alloys
23
Maiiufaciure. — The production of these alloys does not
call for any special care. Like lead, the tin is added when ready
to cast, a short time, before .removing the crucible from the fur-
nace.
Properties. — Brasses containing more than 4 per cent of
tin can no longer be used, being extremely brittle and having a
fracture exhibiting large grains and a silver-gray appearance.
Diagram 4 shows graphically the results of the testing of some
bars which were rolled, drawn and reheated.
These results show (i) that tin increases a little the tensile
strength, (2) that it increases a little the elastic limit, (3) thatiit"
rapidly decreases the elongation and reduction as soon as it
35
So
25
2o
AS
5
N
, '
N
\
>
Diagram 4.
0.7 131 ZJI
Tin Brass. Rolled, Drawn and Annealed Bars
exceeds i per cent, (4) that it produces a very marked brittleness:
and (5) that it increases the hardness. Their deduction agrees
with the appearance of the microstructure of these alloys, and it
follows that large proportions of tin should not be introduced in
brasses. If the brass is to be worked hot, the tin content should
not exceed 2.5 per cent.
An important property due to the introduction of tin has.
been shown by numerous experiments, that of greatl}^ increased
resistance to corrosion by sea water.
Uses. — The composition generally adopted for naval uses
is as follows: copper, 60 to 62 per cent, tin i to 1.5 per cent and
zinc 39 to 37 per cent.
24
The Iron and Steel Magazine
Tin is also utilized in most brasses of great strength in which
it is present in amounts varying between .3 and 1.5 per cent.
Manganese Brass. — Theoretical Study. Mr. Guillemin has
Fig. 20. Cu, 59%; Zn, 40%; Mn, 1%.
Magnified 50 diam.
Fig. 21. Cu, 54%; Zn, 40%; Mn, 6%.
Magnified 50 diam.
Fig. 22. Cu, 65%; Zn, 30%; Mn, 5%.
Magnified 200 diam.
Fig. 23. Cu, 60%; Zn, 30%; Mn, 10%.
Magnified 200 diam.
published several photomicrographs of manganese brass. Al-
though he does not indicate the chemical composition of these
alloys, the microstructure of ordinary brass is easily recognized,
Copper Alloys
25
as further shown in Figs. 20 to 23. It will suffice to note the
structure of the brass containing 54 per cent copper and 5 per
cent, manganese and to compare it with that of an ordinary brass
containing 54 per cent copper and 46 per cent zinc. It would
seem, therefore, that manganese merely takes the place of zinc,,
producing a similar effect. The results shown in diagram 5
were obtained by testing bars of brass containing originally
54 per cent copper and 46 per cent zinc and in w^hich some of the
copper was replaced by manganese.
To confirm these results two allovs were examined contain-
bo
55
So
HS
HO
3S
So
US
2o
^5
^0
S
"^ 1 ^^.— ^_^_^ ■ — — ^ . —
\
II I I
i_6 \pu 4.38 6 6,9s 9,'5 9,53
Diagram 5. Manganese Brasses. Rolled, Drawn and Annealed Blil;
ing (i) 45 per cent copper, 10 per cent manganese and 45 per
cent zinc, and (2) 55 per cent copper and 45 per cent zinc. They
were found to have respectively the same structure as brass
containing (i) 45 per cent copper and 55 per cent zinc and (2)
55 per cent copper and 45 per cent zinc.
Manufacture. — The first samples of manganese brass were
obtained by Stirling and Parker, who reduced some oxide of
manganese by means of carbon in presence of copper and fol-
lowed by the addition of zinc. In 1876 Parsons, who was the
first to study these alloys, added to the copper some ferro-
26
The Iron and Steel Magazine
manganese and utilized the resulting ferro-cupro-manganese for
the preparation of manganese brass and manganese bronze. At
50
H5
Ho
is
3c
25
Zo
15"
'' "
,--'
h.
■*• ^
_^^^->
<*
>
\ *•
/
■"""■
'
\
N
N
N
X
>
0 s -^
7-
Diagram 6. Manganese Brass. First Type, Cast Bars
0,5 1
2 3 S ?>
Diagram 7. Manganese Brass. Second Type, Cast Bars
present cupro-manganese, an alloy of copper and manganese,
is universally employed. It is added in suitable proportions a
Copper Alloys
27
:?hort time before casting. If a large amount of cupro-man-
ganese is to be used, most of it should be melted in the crucible
itself, but a little should be kept to be thrown into the crucible
when ready to cast. It should be borne in mind that manga-
nese may play in the manufacture of brass (as in that of bronze)
a double part: (i) it is a deoxidizer which will reduce the oxides
present in the bath* and '(2) it confers, when present in excess,
special physical properties as shown below.
Mechanical Properties. — The results obtained by the test-
ing (i) of cast bars and (2) of bars which were rolled, drawn and
reheated, are shown in diagrams 5, 6 and 7.
The following results were moreover obtained by Mr. Guille-
min, with some manganese bronzes:
Description
Treatment
Soft Mn bronze
Medium hard M n
bronze (Roma)
Hard Mn bronze
( Roma)
Very hard Mn bronze
(Roma)
Medium hard Mn
bronze
Medium hard Mn
bronze (Roma)
Cast in sand at 1200'' C.
Cast in sand at 1 250° C.
Cast in sand at 1300° C.
Cast in sand at 1400" C.
Cast in sand at 1 250° C.
Cast in sand
Cast in sand at 1250° C.
Hot rolled and an-
nealed
Hot rolled, stamped
and annealed
Hard ferro Mn bronze Cast in sand
Tensile Strengtli
Elongation
tons per sq.
mm.
%
36
14
34
13
31
12
24
9
37
J7
40
20
45
24
45
18
56
30
36
17
Reduction
10
35
45
13
25
i8
15
30
25
10
From these numerous tests the following inferences may
be drawn regarding the effect of manganese:
I. It increases decidedly the tensile strength.
* This effect, however, is open to discussion, especially in the case
of bronze.
28 The Iron and Steel Magazine
2. It increases the elastic limit.
3. At first it increases and then decreases the elongation
and reduction. This increase of elongation produced by a small
amount of manganese is not evident in the case of rolled bars
because a large amount of manganese was at once added.
4. It increases the brittleness but only when it exceeds 4
per cent.
5. It increases the hardness very slowly.
In Mr. Guillemin's experiments the temperatures were
measured by means of a Ducreted optical pyrometer.
In the case of rolled or forged bars, which are extensively
used in the arts, manufacturers are generally able to guarantee
the following properties:
Castings
Drawn bars .
Forged parts
Rolled sheets
Fine wire . ■ •
Tensile
Strength
Elastic
Limit
Elongation
35 to 40 kg.
15 to 20 kg.
15 to 25%
40 to 50 kg.
20 to 30 kg.
15 to 25%
40 to 45 kg-
20 to 25 kg.
20 to 30%
38 to 45 kg.
15 to 20 kg.
i8t0 25%
up to 105 kg.
85 kg.
25%
The properties of drawn or forged bars depend primarily
on the amount of cold work they have received and this is, of
course, also the case with wires.
Some weldless tubes are manufactured of manganese brass
which have a tensile strength of 45 to 60 kg. per square milli-
meter, an elongation of 20 to 40 per cent and a reduction of 15 to
25 per cent. The following two types of manganese brass are
most employed:
1. Cu, 59 to 60 per cent; Mn, traces to i per cent; Zn, 41 to
40 per cent.
2. Cu, 58 to 59 per cent; Mn, 1.8 to 2.2 per cent; Zn, 40 to
39 per cent.
It should be added that brasses containing manganese only
are not those that are most used.
The Manganese Bronze and Brass Company has obtained
the following results in comparing the hardness of some metals
to that of ordinary manganese bronze and to that of manganese
bronze cast under pressure. The test consisted in producing the
same indentation by means of a knife, the pressures required
being taken as a meastire of the hardness.
Metallography Applied to Foundry Work 29
Pressure
Gun metal 12
Wrought iron . 15
Soft steel 20
Soft steel quenched in oil 25
Manganese bronze 20
Manganese bronze hardened by pressure 22 to 23
From these results the Manganese Bronze and Brass Com-
pany recommends the use of these alloys for parts of machinery
which must resist high internal pressure, such, for instance, as
liydraulic cylinders and others.
It is well to remember that manganese brasses are generally
sold as manganese bronzes, although they frequently contain
little or no tin.
{To he concluded)
METALLOGRAPHY APPLIED TO FOUNDRY WORK*
PART II
By ALBERT SAUVEUR
Development of the Structure of Polished Samples of
Cast Iron
\ XT' HEN samples of gray cast iron, polished as described in
the first installment of this article, are examined through
the microscope, numerous small, irregular cavities are revealed
which mark the spaces once occupied by the small particles of
graphitic carbon always present in this grade of iron. Most of
these graphite particles are removed by the polishing operation,
but the small cavities which remain indicate accurately their
former location and shape. Fig. i shows under a magnification
of 56 diameters the appearance of a sample of gray cast iron
after polishing. The irregular cavities just referred to, whether
they still contain their graphite or not, appear as so many black
areas. It will readily be inferred that the appearance of the mag-
nified image of this polished sample of gray cast iron should con-
vey at least as much information concerning the physical and
•chemical characteristics as the examination of the fracture; or,
in other words, that these properties must be closely related to
* " The Foundry," June, 1905.
30
The Iron and Steel Magazine
the number of the graphite particles revealed by polishing, to
their size and shape, their distribution, etc. I shall have occa-
sion to show that very valuable information may indeed be
obtained from the microscopical examination of polished samples
of cast iron without subjecting them to further treatment, but
it will be noted that in these samples the structure of the metallic
part is not revealed. The polishing operation has imparted the
same appearance to the various constituents of which this metal-
lic mass is composed; they have all assumed a mirror-like aspect,
reflecting the light to the same extent, so that it is not possible
to distinguish them from each other. In order to make these
various constituents visible under the microscope it is necessary
Fig. I. Gray Cast Iron, polished
but not etched. Magnified 56
diameters.
Fig. 2. White Cast Iron, polished
in relief. Magnified 100 di-
ameters.
to impart to them unlike appearances through the action of
certain treatments affecting them differently. These treatments
generally consist in subjecting the polished samples to the action
of acids or of some other reagents which attack certain constitu-
ents to the exclusion of others or with varying degrees of inten-
sity; they are generally known as " etching treatments."
Polishing in Relief. — In the case of white cast iron two con-
stituents are present (to be described later) which differ much
in hardness, and if the tripoli and rouge polishing be continued
for a sufficiently long time, and especially if it be conducted on
a soft, yielding backing, a pronounced relief effect is produced
resulting from the greater wearing of the soft constituent. The
Metallography Applied to Foundry Work 31
difference in level of the two constituents obtained in this way
differentiate them under the microscope without further treat-
ment.
Fig. 2 shows the microstructure, magnified 100 diameters,
of a sample of white cast iron polished in relief. The presence
of two constituents is clearly brought out. The soft constitu-
ent appears dark because, being somewhat depressed, each par-
ticle of it pertains, microscopically speaking, of the nature of a
shallow cavity. The differentiation is further assisted by the
soft constituent assuming a mottled appearance, while the hard
component retains its specular aspect.
While such relief polishing makes it possible in the case of
white cast iron to observe some features of the structure without
further treatment, and while it is occasionally valuable, it seldom
reveals structural details which are not better brought out by
an etching or some other developing treatment.
Etching Methods. — Many reagents have been recommended
for etching polished samples of iron and steel, but I shall only
describe here those treatments which, so far as my experience
goes, yield the best results.
The structure of cast iron may be made apparent by etching
polished samples with one of the following solutions : (i) Nitric
acid in absolute alcohol, (2) picric acid in absolute alcohol, (3)
concentrated nitric acid and (4) tincture of iodine. I have named
them in the order of my preference.
Etching with a Solution of Nitric Acid in Alcohol. — A solu-
tion should be prepared containing 10 per cent of concentrated
nitric acid (1.42 sp. gr.) and 90 per cent of absolute alcohol. A
small amount of this solution should be poured in a small beaker
or dish and the polished sample immersed in it for a very short
time (seldom exceeding 10 seconds). It is generally better
not to leave the sample in the solution for more than five seconds
and to repeat the treatment if it be found that the etching w^as
too slight. When the sample is taken from the etching bath,
it should be quickly washed in alcohol and carefully dried, prefer-
ably by means of an air blast, followed by gentle wiping with a
soft cloth; or, lacking a blast, altogether with a soft cloth.
A piece of chamois leather may also be used to advantage
for wiping the specimen, after drying it, immediately before
examining it under the microscope. To that effect it is con-
7,2 The Iron and Steel Magazine
venient to nail a small piece of the leather on a piece of wood and
to rub the specimen over it once or twice. This block should be
kept carefully covered to prevent any dust from settling upon it
The etched specimen is now ready for microscopical exami-
nation.
' Etching with a Solution of Picric Acid in Alcohol. — A solu-
tion should be prepared containing lo per cent of picric acid and
90 per cent of absolute alcohol and the etching conducted exactly
as described for the treatment with nitric acid in alcohol.
Etching with Concentrated Nitric Acid. — The polished
samples should be dipped in concentrated nitric acid (1.42 sp,
gr.) and imimediately held under an abundant stream of running
water. When iron is immersed in concentrated nitric acid it
assumes what is known as the passive state, i. e., it is not affected
by the acid. As soon as the layer of concentrated acid which
covers the polished surface, however, is diluted by the running
water, it attacks tlie iron vigorously, but for such a short time
(since the water soon removes all traces of acid) that there is
very little danger of etching too deeply. One such treatmjent
is generally sufficient to bring out the structure sharply and
clearly, but if the sample be found insufficiently etched the etch-
ing should be repeated in exactly the same manner.
Etching with Tincture of Iodine. — Some tincture of iodine,,
such as may be obtained from pharmacists, should be diluted
with the same amount of alcohol. A little of this solution should
be applied to the polished surface, conveniently, by dipping a
finger's end in the tincture and gently rubbing the specimen and.
repeating the treatment until the surface appears dull or slightly
tarnished. The sample should then be washed in alcohol and
dried.
Heat Tinting Method. — This method consists in heating
the polished sample gently and gradually in contact with the
atmosphere, by holding it over a Bunsen flame, for instance, or
placing it on a hot plate or in some other suitable manner. The:
different constituents assume, in rapid succession, but with
varying velocities, different shades due to the formation of light
films of oxides, in such a way that at no instant of the heating
are two components colored alike. This .method has been ap-
plied extensively and with much success to cast iron by Mr. J. E.
Stead, and I shall have occasion to again refer to it.
.4 New Hardening Furnace
U
A NEW HARDENING FURNACE
THE half-tones, Figs, i and 2, illustrate a new hardening
furnace invented by S. N. Brayshaw, of Manchester, Eng-
land, and which makes use of a salt bath for heating the work,-
special provision being made for controlling the temperature of
Fig.
Hardening Furnace
the bath as well as the whole interior of the furnace and for hand-
ling the material to be hardened.
The inventor's contention is that the treatment of steel
cannot be accurately regulated in an open furnace, and that it
* " American Machinist," March 16, 1905.
34
The Iron mid Steel Mamzine
is necessary to immerse the steel in a liquid to be sure of its tem-
perature to within a few degrees. Having experimented ex-
tensively with lead, he considers that a salt bath is better, and
has brought out a special preparation to be used as a bath in
this furnace. The operation of the latter is based upon the
laws controlling the heating and cooling of steel, and in this
connection two heating and
cooling curves taken by Mr.
Brayshaw are of interest:
One of these curves. Fig. 3,
marked '' S. & D. Best Cast
Steel," is for ordinary car-
bon steel, and has been
taken by drilling a hole
into the piece of steel and
inserting the pyrometer
into this hole, the wires of
the pyrometer being con-
nected with the self-record-
ing indicator. The piece of
steel is then heated slowly
and uniformly in an electric
furnace. During the period
of heating there is an arrest
in temperature at about the
point 726° C. This arrest
in temperature always oc-
curs at the same point for
the same steel. Of course
it is understood that al-
though the temperature is
not rising at this point the
steel is absorbing heat from
the furnace at a uniform rate all the time. After the absorp-
tion is com.plete the temperature again rises regularly as before.
After the temperature has risen for some time the current is
switched off and the steel is allowed to cool with the furnace.
After cooUng with a fair degree of uniformity for some time,
it will be observed that the fall in temperature of the steel
ceases, although the cooling of the furnace and of the steel
Fig. 2. Showing Interior of Furnace
>
.1 .Vi^a' Hanhiiing Furnace
35
itself must be proceeding regularly. Not only does the tem-
perature of the steel cease to fall, but it actually rises. On
the curve in question the rise is only about 2 degrees. In
some steels this rise is as much as 30° C. After the tempera-
ture has risen and rested for a time, it again begins to fall as
shown on the curve. The point where sensible cooling ceases
and where there is an actual rise in temperature is the point of
recalescence, and it will be noted that this point on the cooling
curve is much lower than the point of absorption on the heating
curve, and at first sight it might be supposed that this is due to
Switclied-oflc9
5
—
--
-
__
--
--
"-
-
-
-
--
--
L^.. '"
! ' ' ' 1 1 1 1
r
! ; ; : ! 1 I 1 L
>
\
+T— +
t
r
—
--
_
—
—
-_
—
—
—
-._
-
\i^\i\n\\U
xd
—
- "^^
\
—
—
- —
—
—
—
—
—
—
—
—
—
—
/ -
^ —
—
- —
—
—
—
- —
—
—
—
—
/
\
t-
^
\
: i
1
/
- ^
,
I
1
J
V
td
\
i
k
(Id
t
\
y
p(
_!___
H
^
s.
~V2)^:?,. Cf-
\
^
•" *
s,
fX
_
_ . ^
_ _
l\
__
_
_.
,
(d
^
^;Z— --
=7()7
-A-
c
C^s
= = ■
i;^-
-7C
8;6-
C
— -
—
(^
• '^'/^ 1 1 1 1 i
— — 1~^ — 1~
_
_!!
N
1
_
-
s
f- :
^H — n
—
—
—
—
f
S
V.
—
—
,
j
—
—
—
— -
—
—
—
N
f;v
— -
—
1
'
\
i
\
/
/
s,
__j
V
/
/
'S
/
^
1
5
/
V
/
5"
J-y
1; L^
.
yVn-v
)L
LJ — U
1 L
-L—
1—
1 — L
I— L
-L
1—
L-L
|.k
;j;41U"
XI iJ '
Fig. 3. Heating and Cooling Curve of Seebohm & Dieckstahl's Best Cast
Steel. i\ per cent Carbon
lag in the instruments on which the curves were taken. It will
be observed, however, that the curves have been taken very
slowdy; and as a matter of fact there is very little lag to be
allowed for. It is in the steel itself that this difference arises.
The other curve, Fig. 4, is of S. & D. N. C. (New '' Capital ")
High-Speed Steel. This curve differs very remarkably from
the curve of the ordinary carbon steel. Both the absorption
point and the recalescence point are very gradual and it is very
difficult to say where they begin and end. Mr. Brayshaw bases
his treatment entirely on the curve. He has found it impossible
36
The Iron and Steel Magazine
to harden steel, no matter how many hours it may be soaked,
at any temperature below the absorption point. For good hard-
ening the temperature must be raised so that the absorption of
heat is complete, but when this has been done the temperature
niay be lowered very considerably so long as recalescence does
not begin — that is to say (referring to the diagram in Fig. 3),
the steel from which this curve has been taken must be heated
to a point above 726° C, but it may then be lowered in tem-
perature to about 710° C, and it will still harden perfectly.
By adopting this method, therefore, of heating carefully to
■J-,
TTl
■ -- - - -- _ -
-
—
—
=2^-5= =
i
/ , >l
u
Minuie<
"
L_
—
t
t_
—
i_
J-
,
I
t-
:z— T=-^
W:^^^^^^^^^^EE.
=
1
Z]
-^ -
i-^i=-^^^-
--
■ _\ \ \y
s<818 i 1
\
M -
r/-
:i^_i
U I
,
--.^^
k=p
^_Z^
tJ —
1- -
,s
S9f-
: 3
<: -
'/
3
fTl —
---
--
— ^
t
---
L.
^^_:7.U.;i:C
T
— X !"
1^4
H
±_.
:=::zE=EE=^;EEEE=E
jii ^
_ A ._.
:.-
1 — |—
"
■
= - ^"^
h -
__
— )~--i-
— 1 —
— zHqi--
-.-U4- ^^y-
1 ^,
/
V
1
f-
i_^ ^
~
V
"
-H '
fc^ — ^
—
-q
—
— ^ Zl
V
x^
—
^
\
'^ S
-
—
-
5 ~ - ^S
p
^*-
tz
r ...
N 1 1 1 1 1 1 1 Nj
Fig. .4. Heating and Cooling Curve of Seebohm & Dieckstahl's N. C.
High-Speed Steel
a point a little above the absorption point and then lowering
the temperature, it is possible to harden steel at a very low point
indeed, which means that the risk of breakage is eliminated and
warping is reduced to a minimum.
To utihze this principle, in the design of the furnace a cru-
cible is included for holding the salt bath, and numerous Bunsen
burners provided at the base for heating crucible liquid and
furnace lining, the temperature being determined by an electric
pyrometer which passes through the furnace cover and rests in
the bath, and is read by means of a Whipple indicator which in
The Economic Value of Cast Iron 37
Fig. I will be seen on the stand at the side of the furnace. By
watching the indicator and regulating the burners the tempera-
ture mav be raised or lowered, or it may be maintained at any
point desired. A counterbalanced tray with perforated bottom,
as in Fig. 2, holds the work, and the latter may be lifted clear of
the bath without changing its temperature, as the whole interior
of the furnace is practically at the same temperature as the salt
mixture.
For handling small pieces rapidly in this furnace a set of
grids or open racks are supplied, which, resting one upon another
on the perforated tray, allow the liquid to reach every part of
the work and bring it to its own temperature.
This furnace has been in operation for some time at the
New York office of Geo. Nash & Co., 224 Pearl Street, who control
the furnace in the United States.
THE ECONOMIC VALUE OF CAST IRON *
THE PRODUCTION OF MACHINE CASTINGS OF MAXIMUM
STRENGTH, UNIFORMITY AND SATISFACTORY
MACHINING QUALITIES
By W. H. PRETTY, Manchester Association of Engineers
TN these days of steel it is possible that insufficient attention
•^ is given to materials of construction which are still very
largely employed, and which, if considered in the light of im-
proved methods of production and operation, are undoubtedly
susceptible of partaking to a greater or lesser degree in the
advantages of scientific manipulation. It is beginning to be
acknowledged that for some purposes wrought iron is distinctly
better than steel, and it must be admitted that cast iron is yet
one of the most extensively used materials in the machine shop,
and yet it is only recently that the foundry and its products
have been given anything like the amount of scientific attention
which has long been accepted in the manufacttire and use of
steel. In a paper recently presented before the Manchester
Association of Engineers, Mr. W. H. Pretty discusses the eco-
nomic value of cast iron in the light of present shop methods,
and some abstract of the paper will be found of interest.
* " Engineering Magazine," May, 1905.
38 The Iron and Steel Magazine
'' The value of cast iron in the arts and manufacttires of the
world is such that it must necessarily take an important place
in the great industrial questions of the day, while the vast
supphes held in store for us by nature, of ores of iron and fuels,
and sources of energy capable of being utilized for the pro-
duction of heat by electrical or other means, are sufficient indi-
cation of its continued use by man in the future. It behooves
us, therefore, as engineers, to give the subject all the attention
we can to develop it to the best of our abihty, and endeavor to
make the founder's work a sure science. The most important
appHcations of cast iron (under which term is included pig iron)
at the present time may perhaps be broadly classed as follows : -
'^ I. Its use in the production of wrought iron, steel and
ingot iron, etc., essentially of a metallurgical nature.
''2. Its use in the service of mankind by civil engineers and
others for structural work, bridges, tunneling, pipe lines for
water supply and other purposes.
''3. The production of castings for the general use of
mechanical and electrical engineers in the construction of
machinery, steam engines and other prime movers, etc.
'' 4. The reproduction of works of art, ornamental iron-
work, and various articles required for domestic and other
purposes in times of peace.
''5. The manufacture of malleable castings.
'' The field covered is a very wide one, Class 2 is gradually
passing into the hands of iron founders possessing blast furnaces,
for obvious reasons, but is still practiced on a small scale by
founders engaged in Class 3, while it would seem that Class 5
is likely to be replaced by steel castings."
The exact definition of the term '' cast iron " is almost as
difficult as the precise meaning of the term '' steel." According
to Professor Howe cast iron is defined as iron containing more
than 2 per cent of carbon; just as he defines high-carbon steel
as iron with a carbon content between 0.30 per cent and 2.00
per cent, and low-carbon steel as iron containing less than 0.30
per cent of carbon. Mr. Pretty is much less definite than this, as
he broadly defines cast iron as iron with the addition of small
quantities of carbon, silicon, sulphur, phosphorus, manganese
and '' anything else it can lay its hands on."
'' Cast iron is frequently described as occupying a place in
The Economic Value of Cast Iron 39
the iron carbon series, but it is something more than this, its com-
position is of an exceedingly complex nature, and it should really
be regarded as the product of a process in which other materials,
in addition to iron, are reduced by the action of fuel and blast
and enter into combination with it. In addition to these we
may have compounds formed without the direct reduction of the
elements, and it is not at all unreasonable to suppose that inter-
change between compounds and elements takes place in solu-
tion at high temperatures analogous to those we obtain in the
chemical laboratory in solutions at normal temperature. It is
customary to-day to judge the suitability of a pig iron for foundry
purposes by the carbon, silicon, sulphur, phosphorus and man-
ganese present. Before long we shall have to pay more attention
to other elements, such as chromium, arsenic, copper, titanium,
etc., and their behavior when in solution or combination with
irons, in c[tiantities so small that we to-day ignore them."
After enumerating the more frequently employed ores of
iron, and the processes used in the production of pig iron, Mr.
Pretty discusses the ordinary methods of grading and classifying
the product. Formerly the old method of determining the
character of the metal by the fracture of the pig was considered
sufficient, but this can no longer be accepted. The increase in
varieties, together with the changes in structure due to the pro-
duction of machine-molded pigs cast in chills, demand some
more precise method, and the use of a combination of chemical
analysis, united with physical and mechanical tests, must be
advocated. Of course an accurate knowledge of the pig is by no
means everything, since the advantages of the most careful
choice of materials may be completely upset in the foundry by
careless working at the cupola, the use of improper fluxing
materials, blast conditions, casting temperatures and other
elements of manipulation.
There is no doubt that by the use of chemical and physical
methods of examination, both as applied to the pig and to the
resulting castings, the conditions giving any desired result may
be discovered, and a repetition of any determinate series of
operations be made to give a desired product.
The present interest in the introduction of rapid machining
processes, including high-speed steels for the cutting tools and
more powerful machine tools in which to do the work, has em-
40 The Iron and Steel Magazine
phasized the demand of the machine shop upon the foundry for
soft and easy-cutting castings. As Mr. Pretty well says, how-
ever, the question arises, Which is the more important factor, the
economic value of the material, or the output of the machine
departments? In many cases the strength is ample and the
cost of machining is paramount, but there are many others in
which the material must take premier position. Mechanical
tests show conclusively that metal which would be hailed with
delight and satisfaction in the shop because of its softness,
may be unsuitable to meet the conditions of stress and wear
which are to come upon the finished product. Mr. Pretty urges
the importance of including among the tests for cast iron that
of its machining quality, by the provision of test pieces to be
subjected to the lathe and the drill, so that information upon this
important class of properties may be included with those of
resistance.
In this connection a form of standard test is given, readily
available for trial in the lathe as well as in the drill-press and
boring mill.
^' To enumerate a few of the questions bearing upon the
manipulation of cast iron throughout its treatment in the
foundry we have:
" The formation of suitable slag in the cupola.
" The behavior of the blast and possible occlusion of gases
in the cupola under bad working conditions.
" The addition of minute quantities of reducing agents,
such as ferro-aluminum, ferro-silicon, ferro-manganese (silicon-
aluminum-manganese alloys), etc., to the molten metal while in
the ladle.
" The pouring temperature and the study of cooling curves
by means of thermometers specially suitable for the work.
'' The behavior of moisture and gases at the high tempera-
tures suddenly brought to bear upon them, and ranging from
15° C. to 1600° C. or higher, and their behavior when in a state
of occlusion, solution or inclosure in a mass of cast iron capable
or incapable of yielding to the pressure produced.
" The occlusion of gases by molten cast iron and their
retention or throwing off as the metal cools down to the solid
state, and the effects of rapid cooling upon this phenomenon.
" The possible liberation of gases within the mass of molten
The Economic Value of Cast Iron 41
metal as it cools down, and due to molecular and chemical
change other than that accounted for by occlusion.
'' The effects produced by various degrees of moisture in the
sand composing the mold upon the metal in contact with it,
and methods of determining the percentage of moisture in the
facing and other sands used in forming the molds. The
shrinkage of cast iron as it cools down from the molten state, and
its sudden expansion when nearing the point of solidification,
afterwards followed by the usual contraction of solids as its
temperature falls.
''•The remelting of foundry irons and the change in char-
acter of the chemical constituents, and means of remedying
this as required.
'^ The diffusion of metals.
" The heat treatment of cast iron and molecular rear-
rangement.
" The decay of cast iron under various conditions of em-
ployment, and means of remedying this.
'' The effect of casting temperatures upon special chilling
work.
" The study of eutectic and other alloys within a mass of
cast iron, and the physical changes which accompany them.
" The various types of furnaces in use for remelting pig
irons for foundry purposes, etc., and the nature of the blast
capable of giving the best results."
In the United States the study of cast iron and of foundry
work has been effectively conducted by the American Foundry-
men's Association, and it is a matter for congratulation that a
British Foundrymen's Association has recently been formed,
since it is only by interested cooperation of the men who are
actually doing the work that the same progress may be expected
in this as has been made in other departments of metallurgical
work.
42 The Iron and Steel Magazine
THE CONSTITUTION OF IRON-CARBON ALLOYS *
STABLE AND METASTABLE EQUILIBRIA IN IRON-CARBON
ALLOYS
By E. HEYN, Charlottenburg
Translated for The Iron and Steel Magazine by MILES S. SHERRILL, Massachusetts
Institute of Technology
{Concluded from page ji8, Vol. IX)
B. The Phenomena Occurring on the Solidification
AND Cooling of Iron-Carbon Alloys
TN this field of investigation, Professor Roozeboom f has ex-
-*- pressed his opinions from the standpoint of the phase doc-
trine. I should like to consider his views a little more closely.
The actual hypothesis of Professor Roozeboom has been made
known to you in the preceding address. J I desire here merely
to point out again the most important points. Professor Rooze-
boom takes as his basis the experimental values of Roberts-
Austen ,§ which, as far as they relate to the case in question, are
indicated by the black dots in Fig. i8. He combines these by
means of the following hypothesis: " In iron-carbon alloys below
approximately 1000°, graphite is the unstable, the carbide the
stable form in which carbon appears. At 1000° (along the line
EF in Fig. 18) the following reaction takes place,
martensite -|- graphite <=± carbide."
Roberts-Austen had assumed in his fifth report that the
branch BD in Fig. 18 corresponded to the beginning of the sepa-
ration of graphite, AB to that of iron, and that the line aBc
expressed the separation of the eutectic mixture of iron and
graphite. Roozeboom thereupon pointed out that, along AB,
not iron but mixed crystals of iron and graphite separate, and
that, therefore, the eutectic is formed not of iron and graphite,
but of mixed crystals a and graphite.
A proof for the fact that the line aBc in Fig. 18 really corre-
* " Zeitschr. f. Elektro-Chemie," Vol. X (1904), 491.
t Zeitschr. fiir phys. Chem.," 34, 437 (1900).
+ " Zeitschr. f. Elektro-Chemie," Vol. X (1904), 489.
§ Fifth Report of the Alloys Research Committee, " Engineering,
210 (1899).
TJte Cotistitiition of Iron-Carhon Alloys
43
spends to a eutectic which is in part composed of graphite has
not been given by Roberts- Austen. Such a proof has up to the
present time not been given, not even by the recent investiga-
tions of Carpenter and Keeling,* although the latter investiga-
tors without more ado follow the opinion of Roberts- Austen.
One is, therefore, justified, till further evidence is advanced, in
expressing doubt as to the correctness of this conception. Under
the microscope, graphite is indeed often enough observed, but
lUOO*
r:^^
-•-(►
3 If §■
"Percent C^RBoN
• CuR.vE. OF T?oBERTs- Austen
o 7
MooipiCfSTlON OF S^rtE PyT^OOzEBOOM
Fig. i8
never anything which resembles a eutectic of graphite and any
other structural constituent.
Roozeboom has likewise accepted Roberts-Austen's view
concerning the significance of the line aBc, originally expressed
by Roberts- Austen. As opposed to this view, stands the fact
that iron-carbon alloys low in silicon and containing up to 4
per cent carbon, as they were used by Roberts- Austen in his
experiments, contain after solidification and cooling no graphite
at all. In order to explain this contradiction, Roozeboom put
* Iron and Steel Institute, May, 1Q04.
44 The Iron and Steel Magazine
forward the hypothesis above mentioned and expressed by the
equation taking place at 1000°,
mixed crystals E + graphite ^ carbide.
According to this, the graphite in contact with the mixed crystals
E, should, at 1000°, changeover into the carbide; so that after
cooling, graphite should no longer be observed, but white iron
free from graphite.
From the standpoint of the phase doctrine, this hypothesis
fulfills all requirements; it does not, however, from the stand-
point of the metallurgist. I expressed my suspicions regarding
this to Professor Roozeboom at the time.
How can the incontestable fact that, under otherwise the same
conditions, rapid cooling opposes and slow cooling favors the
separation of graphite, be combined with the above hypothesis?
The entire industry of chilled castings depends on this phenom-
enon. Any explanation of it by means of Roozeboom's hypoth-
esis is so far fetched that really nothing remains of it.
I should like to call your attention to another possible ex-
planation of the solidification phenomena of the iron-carbon
alloys which can be brought into agreement with metallurgical
experience on the one hand and with the solidification curve of
Roberts- Austen on the other. Of course, I do not pretend to
advance here a complete theory; the experimental foundation
for such seems to me to be still much too insecure.* I merely
wish to emphasize again that supercooling phenomena, that is,
unstable equilibria, must be employed more than ever before to
explain metallurgical processes, and that it is not necessary to
have the phase doctrine include everything. My considerations
proceed from the opinion generally accepted in metallurgical
circles, that for solidified iron, graphite is the stable form in which
carbon appears, and the carbide is a less stable, unstable, or
metastable form. This conception is based on the fact that it is
possible by annealing white cast irons which are free from graph-
ite, to separate the carbon in graphitic form (temper carbon).
Charpy and Grenet f have stated this explicitly without refer-
* Even the above-cited experiments recently carried on by Carpenter
and Keeling are by no means extensive enough to render this foundation
more firm.
t Bull. Soe. d'Enc, p. 399 (1902).
The Constitution oj Iron-Carbon Alloys
45
ence, however, to the data of Roberts-Austen or Roozeboom's
hypothesis. I agree with them entirely. My conception, which
I have made provisionally, is reproduced in Fig. 19.
The black full lines ABDa^c indicate the stable system iron
+ graphite. This system tends towards supercooling; the
stable equilibrium, corresponding to the two extreme phases
iron and graphite, is reached only very gradually. The solidifi-
cation can take place in the supercooled, that is, metastable con-
A= 1600
^CtR/VPHiTE
Percent Carbon |
3tABLE Eq\/»HBR»UM , ^RON +GfR^1'H»TE
Me-TASt^ble tqoiLiBFiiuM ; Iron ^ Garbide
Fig. iq
dition, according to the lines Aa'BT, etc. That is to say, pure
iron crystals do not separate out of the liquid alloys, but mixed
crystals of iron and carbide; neither does the eutectic B form,
but the alloy, as a result of supercooling, is capable of remaining
fluid below aBc; the separation of graphite is prevented. The
supercooled alloy enter^the lower region in which the carbide
can exist in a metastable condition. The entire solidification,
therefore, takes place without formation of graphite and with
the separation of mixed crystals a' and carbide. The reactions
46 The Iron and Steel Magazine
taking place on further cooling are easily explained by means of
Fig. 19.
It is probable that the possibility of supercooling exists
only up to a certain maximum carbon content, which corre-
sponds to some point lying not far to the right of B. If the
carbon content of the alloy is greater than this, then along BD
graphite separates out of the liquid alloy and, acting as a nucleus
for further crystallization, frustrates the supercooling. Such
alloys occur in technical use only exceptionally. The majority of
our technically important irons lie to the left of "the point B. It
is a known fact that it is possible to solidify cast irons low in
silicon with 3.5 to 4 per cent carbon entirely w^hite, that is, with-
out any separation of graphite. Here, accordingly, the tendency
towards supercooling, and therefore the tendency to go over into
a metastable condition, is extraordinarily marked. The presence
of silicon decreases the tendency towards supercooling; such
irons with a sufficient quantity of silicon approach more and
more the stable solidification, with separation of most of the car-
b>on as graphite, and the retention of small amounts of carbide
in a metastable condition. If the silicon content be correctly
regulated in relation to the other constituents of the cast iron, it
is possible, depending on how rapidly or slowly the solidifica-
tion is allowed to take place, to obtain white cast iron in the
metastable condition free from graphite, or gray cast iron in
which some graphite has separated, but which partly retains
carbide in the metastable condition. This is, in fact, the funda-
mental principle of the production of chilled castings.
Manganese would favor the tendency to supercooling.
It remains still to be explained why the condition corre-
sponding to the two phases, iron and graphite, has been accepted
as stable. If white, that is, metastable solidified iron, be heated
to higher temperatures, — for example to red heat, — and if at the
same time, what practically is very hard to carry out, the heating
is so conducted that the carbon content remains as far as possible
unchanged, a decomposition takes place in the interior of the
iron. Carbon separates out in a form which is known to the
metallurgist as temper carbon. This kind of carbon can neither
metallographically nor analytically be distinguished from graph-
ite. It is extremely closety related to graphite, possibly even
identical with it. Figure 20 shows an iron originally white,
The Constitution of Iroji-Carbon Alloys
47
tree from graphite, in which this separation has taken place.
The iron contained, before tempering, 2.49 percent carbon, and,
indeed, in a form other than graphite. After annealing for 108
hours,* it contains 2.16 per cent total carbon, of which a large
proportion was in the form of
temper carbon. Figure 20 shows
to the right the dark tem-
per carbon (graphite) imbedded
in a light surrounding of ferrite.
At some distance from this, lies
well-defined pearlite, and in this
to the left, cement ite. The es-
sential point to be noted is that
the temper carbon lies in the
neighborhood of the ferrite. At
the places, therefore, where the
stable condition has been reached,
the two phases iron and carbon
are in contact with each other,
and pearlite still remain unchanged next to each other; there,
the stable condition has not yet been reached. The occurrence
of the stable condition, therefore, starts from single centers,
nuclei, and spreads gradually out from them. This is a sign
that the stable equilibrium is reached only very slowly.
The separation of carbon as a result of tempering is a phe-
nomenon similar to the annealing of steel; only the degree of
instability in which white cast iron, free from graphite, exists is
much less than that of quenched steel. The instability is, how-
ever, sufficient to cause the metastable condition on heating to
change to the stable condition, even at temperatures essentially
below the line a''Bc, and, indeed, the greater the degree of insta-
bility, the'lower will be the temperature at which this change can
take place. Since the presence of silicon works in opposition to
the supercooling, that is, increases the degree of instability of the
metastable condition after it has once been formed, it is to be
expected that the higher the silicon content, the lower will be
the temperature necessary for the separation of temper carbon
from white cast iron. This actually follows from the experi-
FiG. 20. Iron 2.49 C. AnneaU-d
108 Hours
At other places cementite
* Experiments by Ledebur, " Stahl u. Eisen," p. 777 (1886).
48 The Iron and Steel Magazine
ments of Charpy and Grenet.* The presence of manganese
increases the tendency towards supercooling, and, therefore has
an effect opposite to that of silicon.
If the stable condition occurs only after complete solidifi-
cation of the supercooled alloy, and therefore below a'B'F, then
no eutectic between iron and graphite (corresponding to the point
B) can be formed, but the two phases form adjacent layers as in
Figure 20. Since with high silicon alloys, which after solidi-
fication contain carbon principally in the stable graphitic form,
no eutectic is to be detected, it is very probable that the forma-
tion of graphite took place only below a'B'F. This formation of
graphite does not take place suddenly at a constant temperature,
as the line a^'Bc would require if no supercooling took place, but
it is extended over a range of temperature, and, therefore, renders
difficult the determination of the solidification curve by pyro-
metric measurements. In this way, points of the line a'BT will
continually be observed and in fact more distinctly the greater
the supercooling. Below this line only will heat effects of a
variable character take place on account of the gradual separa-
tion of graphite. Only when the supercooling between a"Bc
and a'B'F is removed, does the temperature return suddenly to
the eutectic line (a"Bc) ; then a distinct point must be detected
pyrometrically, and also a distinct eutectic structure between
iron and graphite observed. Whether this case has ever been
observed is not known.
If the formation of graphite take place below a'B'F, then
the longer time allowed for cooling from a'B'F to a temperature
where the formation of graphite is very slight, the more com-
plete ought to be the separaton of carbon as graphite.
The separation of temper carbon in heated white iron is
analogous to the process of " devitrification "; while with glass
the amorphous condition is retained as a result of the super-
cooling, and yields to the crystalline form on devitrification ; in
the case of iron, the supercooled, met ast able condition as well as
the stable condition is crystalline. On de vitrifying glasses,
often together with that of raising the temperature, a
mechanical effect comes into play at this temperature. This can
be observed on drawing out and bending glass rods. A similar
* Loc. cii.
The Coiistitutiou of Iron-Carbon Alloys
49
phenomenon has been observed with highly carburized steels
which have solidified free from graphite. On forging at a red
heat, they show a tendency to pass over partly into the stable
condition by separating carbon, that is, figuratively expressed,
to devitrify. In Fig. 19 the stable system is assumed to be
represented by the equilibrium between the phases iron and
carbon. It appears as if this completely stable equilibrium
occurs only locally in irons, for example as is illustrated by
Figure 20, and as is
also shown in Figure
2 1 . The latter represents a
Swedish charcoal gray cast
iron w4th 3.7 per cent total
carbon. The graphite is
here imbedded in ferrite.
The rest of the structure
consists of pearlite.
The metastable solidi-
fication, see Fig. 19, shows
a eutectic point at B'. The
solidification of irons with a
higher percentage of carbon
than that corresponding to
the point a' must take place
in such a way that at first mixed crvstals a' separate, that on
further cooling the mixed, crystals absume the composition a'
while the mother liquor reaches the eutectic composition B'.
At the temperature corresponding to the line a'B', the
metastable solidification is at an end; the
solidified iron consists of crystals a' and
eutectic B', which is itself made up of mixed
crystals a' and carbide crystals F. Just be-
low the eutectic temperature, therefore, a
structure like that indicated in Fig. 22 is to
be expected. The black places correspond
to the carbide. Since the line a'P slants
towards the left, on further cocking the
quantity of carbide must increase at the
expense of the mixed crystals. The carbide forming will pre-
sumably deposit on the carbide crystals already present, and
Fig. 21.
Swedish Charcoal Gray
Cast Iron
CtnENTlTE
Fig. 22
The Iron and Steel Magazine
therefore, at a temperature just above r, the structure will be
similar to that indicated by Fig. 23.
As soon as the temperature r is reached,
the mixed crystals change over into pearl-
ite. In Figure 25 such a structure
as is regularly observed in high carl:)on
irons can be easily recognized. The cem-
entite is light colored, the pearlite appears
darker and shows the composite laminated
structure. In the middle of the cement ite
are scattered about tiny islands of pearlite,
just as would be expected from our con-
Similar pearlite inclosures can be detected in
The above-
vYhjte: MiXED Crystals
Fig. 23
siderations.
Figure 8.
mentioned conception of the
reactions which take place
on the solidification of iron-
carbon alloys explains quite
naturally and readily the
present' metallurgical pro-
cesses. Whether it is correct
or not can only be shown
by further conscientious ex-
periments. It is to be dis-
tinguished from Rooze-
boom's idea essentially in
the fact that I consider the
transformation
Fig. 25. High Carbon Steel
CeheNtitE
martensite 4- graphite ;=± carbide
not to be supported by experimental data, and therefore as
unnecessary.
Metallographic results directly
contradict this transformation. If a
cast iron in which both graphite and
cementite occur, be considered, then,
if according to Roozeboom cementite
is formed, as in the reaction above,
from martensite and graphite, it is to
he expected that graphite and cementite will lie directly next
to one another as in Fig. 24; that is, a kind of contact meta-
GtRftPHlTE
Fig. 24
The Constitution oj Iron-Carbon Alloys
51
morphosis would occur. This, however, never occurs. The
graphite always lies as far away from the cementite as is possible,
as is distinctly seen in Figures 26 and 27. Both pictures
represent a mottled cast iron. In the former, the graphite ap-
pears dark, imbedded in a finger-shaped part of the structure
which is shown by Figure 27 to be pearlite. The raised little
islands, located at a considerable distance from the graphite,
are cementite, which in their turn, as already mentioned above,
inclose tiny spots of pearlite.
It remains still to be pointed out that in Fig. 1 1 the exact
positions of the individual parts of the diagram have not been
firmlv established ; it is supposed to represent the essence of the
matter only qualitatively and not quantitatively. The two lines
"■ ^ -^ c-\ r>
-?V;^-==!li?'*'^
Fig. 26. Mottled Cast Iron
Fig. 27. Mottled Cast Iron
a"Bc and a'BF may possibly lie very close to one another. It
is undetermined whether a''Bc in Fig. 19 corresponds to the
line aBc in Fig. 18; that is, whether that which Roberts-Austen
observed really corresponded to the stable condition. Corre-
sponding to this case, Fig. 19 is drawn with reference to Fig. 18.
This is probably not the case. It is more probable that a''Bc in
Fig. 19 was passed by in the observations by Roberts- Austen,
and that he in reality observed the line a'B'F (Fig. 19) and gave
it a false signification, and that, therefore, both lines a''Bc and
A'B'F in Fig. 19 should be moved up by a certain unknown
amount.
Further investigations concerning the solidification reactions
are of the highest importance for the foundry practice. It is, for
5 2 The Iron and Steel Magazine
example, still an oijcn question whether, in irons containing silicon
whose tendency towards supercooling has been decreased, and in
which, therefore, partly metastable, partly stable, reactions take
place, the latter already take place while the liquid mother liquor
is still present, or whether the appearance of the graphite forma-
tion with partial replacement of the metastable equilibrium
results below the line a'F only. It is also not impossible to have
both reactions taking place simultaneously.
I
ABSTRACTS
#
{Fro)n recent articles of interest to the Iron and Steel Metallurgist)
Tj^XPERIMENTS on the Fusibility of Blast-Furnace Slags. O.
^^ Boudouard. Iron and Steel Institute, May, 1905, meeting.
20.000 w., illustrated. — The
author conducted an exten-
sive series of experiments
dealing with the fusibility of
blast-furnace slags which are
described at length in his
paper. His conclusions are
as follows :
'' I. The softening temper-
atures of slags which I have
determined during the course
of these investigations may be
regarded as practically their
temperatures of fusion. Nu-
merous observations clearly
show that no greater differ-
ence exists between these two temperatures than exists between
any temperatures of observation referred to actual temperatures
below 1500° C.
'' With regard to the difference existing between the tem-
perature of formation of a slag and that of its fusion, it is, accord-
ing to my observations on the dicalcic silicate and on the
* Note. The publishers will endeavor to supply upon request the full
text of the articles here abstracted, together with all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60
cents, "D" 80 cents, "E" $1.00, "F" $1.20, "G" $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cases
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
ing upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.'
53
54 The Iron and Steel Magazine
aluminate of calcium AUOg, i.SCaO, equalh^ permissible to regard
it as no greater than the amount which would fall within the
permissible limits of probable error. The conditions under
which I worked, i. e., with intimate mixtures of finely powdered
constituents, were highly favorable to this conclusion. In blast-
furnace practice when the fluxes are added in more or less large
pieces, and more or less well mixed, the temperature of forma-
tion of the slag is certainly higher than that of its fusion, but it
will the more closely approach it the more intimate and complete
the previous mixture of the materials has been. For a given slag
the temperature of fusion that I obtained can, therefore, be
regarded as the minimum temperature of formation; and in
ordinary blast-furnace practice there is certainly an economy
of fuel to be realized if the fluxes added have been previously
crushed and mixed.
'^2. I have determined the fusibility curves for the silicates
and aluminates of lime at the same time that I completed those
for the silicates of alumina. The fusion point of silica is equal
to 1830° C.
* ' The examination of the fusibility curves of silicates and
aluminates of lime show that the addition of a small quantity
(about 10 per cent) of silica or of alumina to the lime is enough
to lower the fusion point considerably. The phenomenon is
less pronounced with the addition of lime to alumina; nearly
40 per cent of lime must be added to cause the mixture to melt
at 1500° C. One finds, in the French translation of Ledebur's
' Treatise on Metallurgy,' * that slags into the composition of
which only silica and lime enter are often infusible. According
to my experiments, slags containing 30 to 90 per cent of lime
have fusion points below or at least equal to 1500° C.
''3. The Existence of Definite Compounds, vSi02CaO,
Si022CaO, Si023CaO; AXfi^CM, Al2032CaO, Al2033CaO, the
preparation and properties of which have already been described,
Tias been confirmed by the indications of the fusibility curves.
''4. A triangular diagram, to which are added numerous
•complementary curves, summarizes the extent of our knowl-
edge of the fusibilitv of alumino-calcic silicates. It focuses
and extends the conclusions arrived at by Akerman and Gredt.
* Paris, 1895, Vol. I, p. 202.
Abstracts 55
'' In a general wav the addition of alumina to a silicate of
lime first raises the fusibility, which ascends to a maximum,
and ultimately lowers it. The more basic a siUcate is, the higher
must its alumina content be to make it fusible.
-Influence of the Index. -The following table gives the
composition of slags at the maximum fusion points:
Percentage Composition
Index ^i57 ^^^ cIo Temp.
0.5 23.6 17-8 58.6 1430^
n = 26.6 40.2
i.o 40.0
2.0 571
30
33-2 1460'
22.7 37-3 1345'
16.2 26.7 1300'
66.7 12.6 20.7 1325'
" Substiiution 0] Alumina for Lime, the Percentage Proportion
of Silica remainino^ Constant. - For siUca percentages higher
than 30.0 the fusibility rises markedly up to 25 per cent of alu-
mina, to fall rapidly beyond this percentage.
- For silica percentages of 30 the fusibility varies but little
up to 35 per cent of alumina. Beyond this it falls rapidly.
- For slags with less than 30 per cent of sihca there is a
notable increase in the fusibiUty up to 50 per cent of alumina.
Beyond this the fusibility diminishes very rapidly. The curves
usuallv present two minima.
''The table below gives the composition of the slags at the
minimum fusion points.
SiO, Al,03 CaO Temp. SiO^ AUO3 CaO Temp.
; 40.0 550 1360° 20.0 22.5 57-5 1425^^
,, 50.0
10. o 7-5 82.0
35-0
45.0 1370° 20.0 50-0 30-0 1450^
1460° 30.0 17.5 52-5 1435°
55.0 1370° 40.0 25.0 35.0 1350°
10. o 40-0 1340°
20.0 20.0 1350°
10. o 20.0 136°°
15.0 32.5 52.5 1380° 50.0 lo.o 40.0 1340
52.5 32.5 1455° 60.0 20.0 20.0 1350
70.0
'' I have given in the following tables the relative proportions
of alumina and of Ume corresponding to the maximum and
minimum fusibihty of slags containing increasing increments ot
silica :
^6 The Iron and Steel Magazine
CaO
SiOg Temp. Min. AUOg CaO Temp. Max. AI2O3
10 per cent 1460° 7.5 82.5 1510° .... 90.0
1350° 37-5 52.5 >i85o° >75.o <i5.o
20 per cent 1465° ,, 80.0 1490° 5-o 75-o
1425° 22.5 57.5 1510° 40.0 40.0
1450° 50.0 30.0 >i85o° >72.o < 8.0
30 per cent i435° ^ 70-o i455° 5-o 65.0
1420° 15.0 55.0 >i85o° >65.o < 5.0
40 per cent 1350° 25.0 35.0 i445° " ^°°
1880° 60.0
50 per cent 1325° 15.0 35.0 1440° ,. 5o-o
>i8oo° >47-5 < 2.5
55 per cent 1300° 17.5 27.5 1425° .. 45-o
>i8oo° >43.o < 2.0
60 per cent 1325° 19.0 21.0 1400° ,, 40-°
>i8oo° >39.o < i.o
70 per cent 1360° 10. o 20.0 1500° " i°-°
„ >i75o° >28.o < 2.0
80 per cent 1500° 7.5 12.5 >i75o° < 2.0 >i8.o
>i7oo° >i8.o < 2.0
''5. The triangular diagram constitutes a perfect chart.
enabUng metallurgists to determine without difficulty the tusion
temperature of a given alumino-calcic silicate; and one of the
important factors in the satisfactory running of a blast furnace
is a knowledge of the degree of fusibility of a slag. The calcu-
lation ot the ore mixture is actually carried out by the stoichio-
metric method. The employment of this method necessitates
a complete acquaintance with the chemical composition of the
slag; the considerations are therefore based upon comparisons.
But as the number of substances that may enter into the com-
position of a slag is considerable, and as the burdens are usually
of a very variable nature, it is in most cases impossible with the
ores and fluxes 'available to reproduce at will a silicate of the
chosen type ; hence it is necessary to rest content with finding a
compound that will approximate as closely as possible to it in
its essential characteristics, i, e., particularly in its degree of
fusibility.
" In metallurgical operations it is very rarely that slags are
produced containing nothing beyond silica, alumina and lime;
there is always more or less magnesia, manganese and oxides of
iron present. Nevertheless it is amongst this class that the
great bulk of slags produced in the manufacture of pig iron falls.
.4 bstracts
57
d.
E
oooooooooooooooooo
OiftOiOOiftOiOiOOOOOiOOlOOiO
<
eoec p 00 «p b-»05 op p«; ricooj ; : ;
©fl W "N 1^ 0 © 0 © -^ »^ i-H 0 1-H 1-1 C5
0 © TT "O* rj«(N'^f^
«2
<
eo© -^ lO
ec« -© --...- .TT
6
c<ic^'«rOi © © ©(M © © i-<eo © ©
cot^rrt- r : r :^'*©t^c<ipt-©ip»p
^^i<\h © © 1^ © i^H c<i © .^ (N c<i
d
c
CO t-H © lO Oi©©C^iO t^QO ©©
OC<«C<l© 4i-©©©© ©M .^M
d
U
^o«ot^^5©■n"t^c<^©©■<»<co©cococ^©©
ip<TfcpC^giprH(rjipC<ir-i»PqOr-(t-rHOp
c^QO»b.^oC(^^w<»oo^-«b5b©t--«b(^ooo«j^
d
IS
00 ■«»« t^ go t^ «o 00 © ©oi ooi(Ncoai
00 p 0 C5 CO r-< © -poOpO pco p rr • -
© C<J © C^ 1^ rH 1^ " ^ C^ ,li CO i^ T^ CI C<J
0:
<
cooo-^t^iOg>i-it^©©'>!r05i©'<9<eo'^
cowpr-(t^pq07-t-c^ppoooqoppp
M«bMC<J<3i«©iX)«boi©Tf"65056i©Tt<<i>
r-l ,_! i-i r-i r-< f-H <N -^ eq ^- r-"
0
CO
OOOiOi©© t-©©©©«Or-.CO<M ■<»< ©
pqpt»co»pcoc^^r-iTfcopp(Npaopp
coiw6it~'-'»^<Nt-ibt»t>.©c<jt^.^AHW
;oo>o-vcocococococoeoc^'*'<»'cocoeoco
B
s
"0
1
2ii:3S2 C 3cOo>5<UO
V ' V ■*
i
M
0
>»
3
M • g • ...
0 CO 0 c
rt . - 0 w 0- -
J t^o 2i f ?
0 0 > U, CQ
58 The Iron and Steel Magazine
'' Using the practical data to be found in metallurgical
treatises (Ledebur, Babu), I have determined the fusion tem-
peratures of industrial slags derived from various sources. They
are as follows:
Percentage Composition
Description of Slag
Ferro-silicons with 14 \
per cent S
Gray hot-blast char- \
coal pig iron S
Gray hot -blast coke )
anthracite and coal >
pig iron )
White pig iron — i .
Charcoal
White iron — 2. Coke <
Spiegeleisen
Si02
A
AI0O3
CaO*
Temperature
25.0
25.0
50.0
1450°
35-0
25.0
40.0
1375'
45-0
10. 0
45-0
1360°
65.0
5-0
30.0
1430°
30.0
20.0
50.0
1420°
3 50
10. 0
55-0
1420°
45-0
10. 0
45-0
1360°
40.0
5-0
55-0
1425°
30.0
10.0
60.0
1440°
40.0
5-0
55-0
1425°
CaO MnO
30.0
10. 0
55
.0 5.0
1440°
40.0
10. 0
10
.0 40.0
1440°
The temperatures given in the last two tables are obviously
only approximate, but if the conclusions of Akerman who
describes the influence of foreign oxides on the fusibility of
alumino-calcic silicates be borne in mind, and it be remembered
that as a general rule a slag is more fusible the larger the number
of constituents it contains, the data given above will, notwith-
standing, furnish information of value to metallurgists, inas-
much as it represents the maximum fusion temperatures of slags
corresponding to a given chemical composition. No. 38a.
The Electro-Metallurgical Industries. J. B. C. Kershaw.
^' The Electrical Age," April, 1905. 4,800 w., illustrated. —
This article includes a short description of some of the most
successful electrical methods for the production of iron and steel.
No. 381. C.
Electric Power in Steel Works. '' The Engineer " (London),
April 28, 1905. 1,200 w., illustrated. — A description of a
number of special applications of electric motors to steel works
machinery, including turning tackle for forging, charging ma-
chines, roller tables and special overhead travelers. No. 382. B.
* Under the term " lime " has been included all the oxides o.her
llian jilica and alumina.
Abstracts
59
An Automatic Stock-Line Recorder for Iron Blast Furnaces.
J. E. Johnson, Jr. A paper read at the February, 1905, meeting
of the American Institute of Mining Engineers. 4,000 w., illus-
trated. — The author describes a patented device consisting of:
(i) a test rod suspended at its top by a chain, but normally rest-
ing on the stock and descending with it except when the bell is
open; (2) a simple mechanism for lifting this rod out of the way
of the in-coming stock at the latter period; and (3) an attach-
ment for recording its motion on a reduced scale. No. 383.
Experiments Relating to the Effect on Mechanical and Other
Properties of Iron and Its Alloys Produced by Liquid- Air Tempera-
tures. R. A. Hadfield. Iron
and Steel Institute, May,
1905, meeting. 35,000 w.,
numerous illustrations. —
From the results of an ex-
haustive set of experiments
dealing with the effect on
mechanical and other prop-
erties of iron and its alloys
produced by liquid-air tem-
peratures the author arrives
at the following conclusions:
'' It is very clear that as
regards iron and iron alloys,
with, however, certain excep-
tions, the effect of low tem-
peratures is to increase in a remarkable degree its resistance
to tensile stress, ordinarily known as the breaking load or te-
nacity, and to reduce its ductility, as measured by elongation,
from the highest point, for example, in mild steel 30 to 40 per
cent, to practically nil. The changes take place to the same
extent, and this is very curious, in the softest wrought iron as
represented by the specimens ' S.C.I.,' ' L.S.S.' (the famous
Swedish melting iron) and also English wrought iron, and in
carbon steel samples from o.io per cent or 0.20 per cent to
the high percentages such as 1.25 per cent or 1.50 per cent.
Thus the absence or presence of carbon in ordinary carbon steel
in which other special elements are not present seems to have
6o The Iron and Steel Magazine
but little influence. That there is no error in this statement is
proved, independently of the tensile tests, by the fact that several
bars of the ^ S.C.I.' and mild steel specimens were submitted to
the low temperature test, and tested by hand hammer imme-
diately after immersion. In all cases they exhibited great brittle-
ness, breaking off instantly upon being struck with the hammer ;
there was an entire absence of ductility.
^' Further confirmation is obtained by the Brinell hardness
ball test, a method of testing explained elsewhere. Under this
test the ' S.C.I. ' specimen at normal temperature had a hardness
number of 90, whereas when tested at about — 182° C. this in-
creased to no less than 266, or about equal to the hardness of
0.80 per cent carbon steel at normal temperature. This almost
seems incredible when it is remembered that the ' S.C.I. ' shows
by analysis 99.82 per cent of iron, and normally has only 20 to
22 tons tenacity with 25.30 per cent elongation.
'' The importance of the discovery of the toughening effect
of nickel upon iron at low temperatures will be seen when it is
understood that whilst it has been well known that nickel in
certain percentages produced important improvements in the
qualities and properties of iron and steel alloys no microscopical
or chemical research work has yet proved why this came about.
To the author it seems clear that these experiments go a long
way towards offering a satisfactory explanation.
''It will be seen that the purest iron, as represented by the
' S.C.I.' containing 99.82 per cent iron and of specially high
quality and purity, becomes brittle to an extraordinary degree
under the influence of the low temperature — 182° C, whereas
nickel itself tested at the same low temperature has improved
rather than deteriorated, not only in tenacity, which iron also
does, but in ductility, in which latter quality iron entirely
breaks down. If nickel, therefore, is present in an iron alloy
containing but little carbon or comparatively low in that
element, it acts as a preventive of brittleness, or is a very con-
siderable modifier of that objectionable quality.
''It may be interesting to state that at ordinary tempera-
tures the toughness or ductility of nickel is no greater than that
of iron. For example, in comparative tensile tests, made by the
writer, of nickel and pure iron, the ductility of iron was greater.
The reduction of area in the material generally shows its condition
A hstracts 6 1
as regards ductility; in the specimens in question the reduction
of area in the tensile test bars was nearly 20 per cent greater for
iron in both the author's ' vS.C.L' and Arnold's pure iron than
in the nickel specimen tested.
'' Iron to a more or less degree, at any rate in manufacturing
operations, always seems to be endeavoring to wander out of
the * paths ' of ductility and toughness; it is constantly endeav-
oring to become brittle. It will often assume its apparently
brittle nature on the slightest provocation, and the metallurgist
by his arts is always trying to correct this tendency. As with
humanity, there seems to be a law of tendencies, and iron by
heredity is constitutionally weak. It would appear, therefore,
that iron, a cheap and convenient metal itself, must be per-
meated by some element that will mask or modify its properties.
Until comparatively recently carbon was the only element known
to modify the properties of iron; but as will be seen in this
research, this element, where great toughness is required, only
helps to make matters worse.
'' Fortunately for iron, however, its close companion, nickel,
singularly enough in the same group, comes along and acts as
a friend in keeping it — iron — up to the mark and preventing
it from wandering out of the narrow road of metallurgical recti-
tude, that is, of toughness or ductility. Exactly why this
should be so cannot easily be explained, but this is the fact.
Possibly some interpenetration of the atomic mass causes a
change which cannot as yet be deduced by any known chemical
investigations. Iron, too, is a very crystalline metal, whereas
nickel appears to be much more amorphous; it is possible,
therefore, that nickel tends to prevent iron crystallizing in this
manner, or prevents it cooling in such large or dangerous type
of crystals. This action of nickel is simply marvelous in
certain of the alloy specimens; for example in an alloy of
iron, carbon 1.18 per cent, nickel 24.30 per cent and man-
ganese 6.05 per cent. Here the ductility is extraordinary at
not only ordinary but low temperatures, probably the highest
known for any iron alloy, and certainly for an alloy having such
tenacity as 84 tons per square inch. There is still present in
this alloy 68 per cent of iron, yet the tendency of the latter metal
to wander into the paths of brittleness is not only entirely checked
at the liquid-air temperature — and this brittleness, as shown so
()2 The Iron and Steel Magazine
clearly in this research, occurs to an extraordinary extent in
pure iron cooled to — 182° C. — but the elongation or ductility
already so great is considerably increased, namely, from 60 per
cent to 67^ per cent. There is also an increase of tenacity in
both cases, namely, a rise of from 10 to 38 per cent. Thus the
nickel present — as these results cannot apparently be ascribed
to any other cause — enables the bar under this high tension and
at — 182° C. to remain far more ductile than the very best of
ductile iron of one third the tenacity. Although the action of
nickel has been specially referred to, it must not be overlooked
that in this alloy there is also present 6 per cent of manganese,
which in its ordinary combination with iron, that is, with no
nickel present, would confer intense brittleness upon the iron
and render it more brittle than if not present. This treble com-
bination of nickel-manganese with iron appears to reverse all
the known laws of iron allo3^s.
^' M. Osmond's theory as regards these iron -nickel-man-
ganese alloys is that manganese acts here in the nature of nickel.
He considers that i per cent of manganese is equivalent in its
action upon iron to 2 per cent of nickel.
" In conclusion, it may be said that the many extraordinary
changes brought about in the physical properties of iron and its.
alloys could not have been deduced from any known laws. Iron
in the main is ' embrittled ' to an extraordinary degree by
liquid-air temperature, and yet it will be seen that this ' heredi-
tary ' tendency can be entirely checked in certain of its nickel-
iron and nickel-iron-manganese combinations.
" These various changes appear to be certainly not chemical,
and it is rather to the physicist we must eventually look for a full
and correct explanation of the many curious results obtained in
this research." No. 384.
The Electrical Driving of Rolling Mills. '' The Iron and Coal
Trades Review," April 21, 1905. 2,000 w., illustrated. No. 385. B.
Microscopic Observations on Naval Accidents. Part II.
Thomas Andrews. *' Engineering," May 5, 1905. 2,000 w., illus-
trated. — A report of the microscopical examination of a portion
from the propeller tail shaft of a large screw steamer, which had
fractured in service. No. 386. B.
Abstracts
63
Accidents Due to the Asphyxiation of Biast-Furnace Work-
men. B. Thwaite. Iron and Steel Institute, May, 1905, meeting.
4,500 w., illustrated. — The
author after describing the
nature and escape of blast-
furnace gas suggests the fol-
lowing precautions in order
to secure comparative immu-
nity from fatalities due to
asphyxiation :
'' I. Where practicable,
all gas flues and gas pipes
should be placed above the
ground, and, if possible, above
the breathing level.
^'2. If the gas flues are
underground they should be
built of blue bricks set in
cement.
''3. Gas-flue manholes should be equipped with lids that
will secure the maintenance of perfect gas tightness.
'^ 4. All habitable buildings should be thoroughly well ven-
tilated. The ground surface below the building should be
covered with a layer of cement concrete, and an open-air space
should be formed between the cement surface and the flooring
boards of the building.
" 5. Forced ventilation should be the principle adopted lor
all occupied buildings in iron and steel works. An air supply
tapped from the cold-blast main may be found useful in this
connection.
" 6. The gas engine exhaust pipes should terminate at a con-
siderable altitude above the works floor, and part of the air
•drawn into the air-tub of the blowing-engine should be supplied
from the engine house.
" 7. Especial precautions should be taken during flue- and
stove-cleaning periods, and in blowing-in furnaces.
*' 8. Enamel lettered warning-notices should be prominently
posted throughout the works.
" g. Fines should be imposed on workmen for gas risks care-
lessly undertaken.
6'4 The Iron and Steel Magazine
'* lo. A paid ambulance corps should be established.
'' II. During the blow-out period of slag-tapping operations
the men should be withdrawn from the zone of gas-escape.
^'-12. The time allowed to be occupied in gas-producer pok-
ing operations for individual men should be limited to three
minutes.
'^13. The reversing valve-pits of gas-fired furnaces should
be well ventilated.
^'14. Roll-call checks of labor employed in gas-flue and
stove-cleaning operations should be established.
''15. If power gas-pipes have to be placed underground,
they should be placed in grate-covered open trenches, and the
pipe-flanges should be planed." No. 387.
Cast-iron Car Wheels. J. E. Muhlfeld. ^' The Railroad
Gazette," May 5, 1905. 2,500 w. — The author's concluding
remarks are given below:
^' In conclusion it would appear that the past and present
undesirable performance of the cast-iron wheel may be overcome :
'' First, by designing a partial or full double plate which
will distribute the metal to provide ample flexibility between
the hub and the tread ; strengthening directly below the rim by
reinforcement of more gray iron to prevent extreme chilling and
cross and longitudinal cracking ; removing the depth of the metal
at the tread and rail contact, which will increase the chill at the
greatest wearing point; increasing the metal at the base of the
flange with more gray iron to draw the chill and strengthen
the throat, and providing a tread and throat contour which will
relieve severe frictional contact at the throat of the flange with
the brake shoe and rail.
^' Second, by the use of a uniformly good material and proper
foundry practices processes and equipment.
"■ Third, by specifications, guarantees inspections and tests
of sufficient severity to iUvSure a proper degree of endurance,
mileage and safety.
'' Fourth, by a flexible suspension of the brake beams that
will prevent the liability of severe friction between the brake
shoe and the throat of the flange, or concentrated pressure on the
rim, either on straight track or when curving.
'' Fifth, by the adoption of simple and substantial anti-
Abstracts 65
frictional side bearings and center plates, lateral motion truck
device and adequate side movement for the couplers the design
of which will not be discussed in this paper although their appli-
cation should be given serious consideration in connection with
the use of either cast-iron or steel wheels.
" SixtJi, by adopting a rail head section that will conform
more closely to the contour of the wheel flange.
*' A construction such as recommended might bring the cast-
iron wheel to such a state of perfection as would make its use
practicable, safe, efficient and economical in connection with the
heaviest capacity freight locomotive tender and car equipment.
^' Considerable thought has been given recently to the
advisability of substituting the steel wheel for the cast-iron wheel.
The value of such a step is as 3^et only problematical. If steel
wheels as manufactured to-day were used in place of good cast-
iron wheels for freight cars the investment in the United States
would be approximately $200,000,000 more than at present, or a
sum sufficient to fully equip 3,500 miles of modern American
railroad. It therefore appears that every opportunity should
be given to perfecting the cast-iron wheel which has given such
good service under the lighter equipment, until it can be made
to meet the present and future requirements or until a steel
w^heel can be produced that can insure equivalent or greater
efficiency and economy." No. 388. B.
R. A. Hadfield's Presidential Address. The Iron and Steel
Institute, May meeting 1905. 40,000 w., many tables and
illustrations. — In this intensely interesting address the author
deals with most of the problems of moment to iron and steel
metallurgists. The address also includes the portraits of many
past metallurgists and scientific men, five of which will be found
reproduced in our frontispiece, as having been more closely con-
nected with the iron and steel industry. They are: Rene
Antoine Ferchault Reaumur (France), 1683- 175 7, physicist;
wrote on cementation and decarburization.
Emanuel von Swedenborg (Sweden), 1688-17 7 2, engineer
and theologian; wrote on iron.
Benjamin Huntsman (England), 1704-17 7 6, manufacturer;
first to melt steel.
Torbern Olaf Bergman (Sweden), 1 735-1 784, chemist;
66 The Iron and Steel Magazine
founder of analytical chemistry, drew distinction and explained
differences between pig iron, wrought iron and steel.
Henry Cort (England), 1 740-1800, engineer; introduced
puddling and the use of grooved rolls. No. 389.
Wrought Pipe Threading and Durability. F. N. Speller.
Read at the annual meeting of the Canadian Mining Institute,
Montreal, Canada, March 1-3, 1905. 3,000 w., illustrated. —
The author recalls the opinion of recent investigators, and nota-
ably those of Prof. H. M. Howe, regarding the relative value of
wrought iron and low carbon steel for the manufacture of tubes,
and describes some experiments conducted in the laboratory of
the National Tube Company, McKeesport, Pa. ''Mild steel,"
the author writes, '' has had to live down prejudice in nearly
everv line where it has displaced wrought iron. The most suit-
able quality of Bessemer steel for pipe manufacture has been the
result of much extensive experimenting and has proven much
more satisfactory in welding. Naturally the makers of wrought-
iron pipe to whom a supply of suitable steel is not available make
the most of the suspicion which the word ' steel ' still raises in
the minds of some people not informed on the modern advances
in this industry. It seems quite reasonable to suppose that the
advantage of superior strength, ductility, homogeneity, finish,
and lower cost, which accompanies the use of steel pipe, would
be more generally recognized if such misapprehensions as these
regarding threading and corrosion were cleared up." No. 390.
Air Blast for the Foundry Cupola. W. H. Carrier. '' The
Iron Age," May 11, 1905. Abstract of a paper read before the
Buffalo Foundrymen's Association, April 18, 1905. 2,200 w. —
The author presents some arguments in favor of the fan over the
positive blower to provide the blast for a foundry cupola. No.
391. B.
Gas Blowing Engines. Tom Westgarth. '' The Engineer "
(London), May 5, 1905. Abstract of a paper read before the
West of Scotland Iron and Steel Institute. 2,500 w., illustrated.
— The author describes some of the best-known gas-blowing
engines, and comments critically upon their performance. No.
392. B.
Abstracts 67
The Fremont Method of Determining the Fragility of Iron
and Steel. T. Y. Olsen. A paper read before the Engineers'
Club of Philadelphia, Pa., December 3, 1904. 3,000 w., illus-
trated. — A critical discussion of the well-known Fremont
impact test. No. 393.
Slag Cement. W. B. Ruggles. " The Iron Age," May 18,
1905. 2,200 w. — ■ The author gives evidences of the high quality
of slag cement, which he claims " compares favorably with the
best brands of Portland cement for at least 90 per cent of all the
work for which cement is used." No. 394. B.
Roll Turning : An Important Mechanical Feature in the Iron
and Steel Industry. W. S. Standiford. '' The Engineering
Magazine," June, 1905. 7,000 w., illustrated. No. 395. B.
METALLURGICAL NOTES AND COMMENTS
The thirty-sixth annual meeting of the
x: Jj T J ° Iron and Steel Institute was held at the
of the Iron and
Steel Institute Institution of Civil Engineers, in London,
May 1 1 and 12. The report of the Council
stated that during the year 1904 there were added to the register
213 names, bringing the total roll of membership on December i,
1904, to 1,910. The following Andrew Carnegie Research
Scholarships, each of ;!^ too, were awarded by the Council: H. C.
Boynton (Cambridge, Mass.), L. A. Guillet (Paris) and W. H.
Hatfield (Sheffield), while further grants of ;^5o each were made
to P. Breuil (Paris), Dr. H. Carpenter (National Physical Labo-
ratory), Messrs. E. G. L. Roberts and E. A. Wraight (London)
and W. Rosenhain (Birmingham).
The retiring president, Mr. Andrew Carnegie, in a very happy
speech introduced Mr. R. A. Hadfield, who, after a brief reply,
presented the Bessemer gold medal to Prof. J. O. Arnold of
Sheffield. Professor Arnold returned his thanks and the presi-
dent then delivered his address.
The following papers were read and discussed:
" The Continuous Steel Process in Fixed Furnaces," by
S. Surzycki.
" Developments of the Bertrand-Thiel Process," by John H.
Darby and George Hatton.
" The Effect of Liquid-Air Temperatures on Iron/' by R. A.
Hadfield, President.
" The Cleaning of Blast-Furnace Gas/' by Axel Sahlin.
" Dry-Air Blast " (Supplementary Paper), by James Gayley.
" Fatigue in Metals," by Sidney A. Houghton.
" The Fusibility of Blast-Furnace Slags," by O. Boudouard.
''Asphyxiation of Blast-Furnace Men/' by B. H. Thwaite.
" Sulphur in Coke and in the Blast Furnace/' by Prof. F.
Wuest and F. Wolff.
An abstracted report of the research work of the
Carnegie scholars during 1904- 1905 was also presented and
discussed It included abstracts from the following reports:
68
Metallurgical Notes and Comments 69
'' Heating and Cooling of High-Speed Tool Steels," by H. C.
H. Carpenter.
" Effects of Reversed Stresses in Steel," by J. C. Gardner.
'' (a) Troostite and (b) Heat Treatment and Fatigue of
Steel," by F. Rogers.
'' Magnetic and Electric Properties of Steel," by Gunnar
Dillner and A. F. Enstrom.
Some of the papers read at this meeting will be found repro-
duced either in fiill or in part in the present issue of The Iron
and Steel Magazine, while the others will be dealt with in sub-
sequent issues.
Practical Micro-Metallography. — Mr. J. E. Stead recently
read two papers before the Royal Microscopical Society of Lon-
don which are briefly summarized in the " Journal " of that
Society for April, 1905.
In illustration of the subject a series ot views were shown
upon the screen, the first twenty of which showed the different
kinds of apparatus used for the preparation and examination of
the specimens. These were followed by a large number of actual
specimens depicted upon the screen in the most brilliant man-
ner by means of the epidiascope, the details of surface and
especially the coloration, being exhibited on a scale and in a
manner impossible by any other means ; the extremely beautiful
colors produced by heating, and especially those upon a polished
section of a meteorite, being amongst the finest examples
exhibited.
Micrograms reproduced from the illustrations in Dr. Sorby's
original papers clearly showed that as far as his work went, it
was of a good and accurate character and that subsequent
observations by more modern workers had confirmed all he had
done.
In referring to the work of Osmond upon steel it was shown
by his illustrations, and also by the work of the lecturer and
other observers, that whilst the hardenite in steels quenched
from a point a little above the recalescence point Ar^.j-g, although
crystalline, was practically amorphous, on heating to and quench-
ing from a higher temperature, a crystalline structure was
strongly developed and had the same characteristics as martens -
ite in steels containing less carbon.
70 The Iron and Steel Magazine
Troostite and austenite, although not thoroughly under-
stood, had been recognized as true micro-constituents.
Sorbite, like troostite, required more study. It was neither
troostite nor pearlite, and Osmond had described it tentatively
as unsegregated pearlite. As much discussion had taken place
during the last few years with regard to the nature of the micro -
constituents, sorbite, troostite and austenite, a committee
had been formed to work under Dr. Glazebrook, of the National
Ph3^sical Laboratory, to endeavor to ascertain their true nature.
The work of Professor Arnold was illustrated by slides made
from accurate drawings ot the structures of cement steels and
steels containing sulphur.
The lecturer expressed great appreciation for Arnold's
drawings, and pointed out that in many cases they were prefer-
able to photographs, but that generally photographs were better
when properly taken.
The special features of Prof. H. Le Chatelier's work were
illustrated by photomicrographs of cast irons and steels, some of
the structures of which had been developed by the action of
potash and lead oxide, which darkened the cementite but left
the other constituents white.
The mid-ribs of cementite in the dark barbs of martensite
were in this way clearly shown.
The effect of strain and continued reversals of stress on iron
was illustrated by photomicrographs prepared by Prof. J. A.
Ewing, Mr. Walter Rosenhain and Mr. Humphrey.
The surface flow of metals was illustrated by the elaborate
work of Mr. G. T. Beilby.
The work of Mr. W. H. Merrett of the Royal College of
Science was represented by photographs of granular pearlite
magnified i,6oo diameters, which showed that the carbide of iron
or cementite existed in globular or roughly-shaped globular
particles completely separated from each other.
The micro-structures of steel produced by electrical pro-
cesses were given by slides provided by Mr. F. W. Harbord, and
shown to be identical with the micro-structures of steels made by
the ordinary processes.
The structure of bronzes was very beautifully illustrated
with slides provided by Messrs. C. T. Hey cock and F. H. Neville,
Cambridge. The lecturer, in describing the work ot'those gentle-
Metallurgical Azotes and Comments
71
men remarked that the research upon copper and tin alloys
presented to the Royal Society was of the highest merit, and a
type of work such as students who wished to study metallic
alloys should take as an example.
The Brinell System of Testing the Hardness of Materials.* —
The method introduced by the Swedish engineer, Brinell, for
testing hardness, consists in measuring the superficial area of
the cavity formed in the sample by a ball of hardened steel (of
the kind used for ball bearings) under the influence of a given
pressure, the ball being half embedded in a backing plate of hard
Fig. I. Press for Pressures of 3 to 6 Tons
steel. On dividing the area of the cavity, expressed in square
millimeters, by the pressure expressed in kilos, the resulting
values give the " index of hardness " of the material under
examination. It is found advisable to employ balls of one given
size (10 millimeters in diameter) in order to obtain comparable
results, and the margin of error may be still further reduced by
making several impressions under different pressures, and select-
ing from these the one corresponding most nearly to a given
standard area as the basis of calculation.
Apparatus for performing these tests is being put on the
market by Mr. H. Huber of Hochst-on-Main. Fig. i represents
* " The Iron and Coal Trades Review," May 5, 1905.
The Iron and Steel Magazine
a press giving pressures of 3 to 6 tons, and suitable for electri-
cians, coppersmiths etc., for testing the hardness of copper
collectors and other fittings. The pressure is recorded on the
dial shown in the head of the press, and the ball has a maximum
stroke of 2 inches. A larger form of the same press is made for
testing objects of greater thick-
ness, in which the stroke of the
ball is 6 inches, the actual height
above the base plate being ad-
justed up to 14 inches, accord-
ing to the thickness of the test
piece, by wheels and locking nuts
on threaded pillars at either side
of the frame.
Figs. 2 and 3 show the new-
est pattern for working with high
pressures (60 to 100 tons). In
this case the pressure apparatus
is utilized direct for setting the
ball at a level corresponding to
the thickness of the sample under
test. The pressure apparatus, Fig. 2, can also be combined
with suitable attachments, such, for example, as that in Fig. 3,
which is used in testing rails.
The measurement of the diameter of the impressed cavity
is performed by the micrometer gauge illustrated in Fig. 4.
Figs. 2 and 3. Apparatus for
Higher Pressures, 60 to 100
Tons
Fig. 4. Micrometer Gauge
This is placed on the test block and leveled by means of the four
set screws until the gauge points just touch the surface of the
material, the fixed point resting exactly on one edge of the cavity
and the other one moved to and fro by turning the micrometer
Metallurgical Notes and Comments
73
screw, until it touches the other edge at a point diametrically
opposite the former one. By this means the diameter of the
imprint can be gauged to within the one one-hundredth part of a
millimeter. A magnifying glass is employed to examine the posi-
tion of the gauge points. By applying the results of this measure-
ment to the apparatus represented in Fig. 5, the depth and angle
of the cavity can be read off at once on the crossed scale without
anv calculation or reference to tables.
Fig. 5. Measuring Apparatus
One of the great advantages of this system of testing is
that it can be applied to any finished article {e. g., projectiles,
steel plates, etc.) without any injury to the latter. On the other
hand it is equally applicable to small test blocks or strips, there
being no distortion of the material when the thickness of the latter
exceeds one tenth of an inch, and the length and w4dth are
over I J inches. Provided the test object is parallel on two sides
and has one plane surface, it need not be trimmed to any par-
ticular shape nor even polished, though, of course, more accu-
rate measurements can be made when this has been done.
74 The Iron and Steel Magazine
In his presidential address, delivered recently
^^offronTrer ^^^^^""^ "^^^^ Institution of Mechanical Engi-
neers, Mr. Edward P. Martin called attention
to a question which, although of supreme importance to the
manufacturing industries of this country, does not appear to
be generally appreciated. This is the rapid exhaustion of not
only our own best iron ores, but also of the most readily avail-
able foreign supplies. So serious a view does Mr. Martin take
ot the outlook in this direction, that he considers it quite worthy
of a Royal Commission for its thorough investigation. Nor is
this view by an}^ means extreme. The country is quite as
dependent upon cheap iron as upon cheap coal for its industrial
welfare. There was a time not so very long ago when the iron
industry was absolutely dependent upon the joint occurrence of
iron ore, fuel and limestone in close proximity to the works.
Dowlais, in these early days, drew its supplies of these three
essential materials from the same hillsides, where works were
constructed for coking the coal, calcining the ore and smelting
the iron. A period of more than a century, however has seen
great changes in these favorable conditions. The Dowlias of
to-day procures its fuel from some dozen miles away, and the
bulk of the iron ore is imported from abroad.
Similar stories of altered conditions and exhaustion of the
best ores can be told of other districts, and in face of this rapid
depletion, the demand for steel and iron is continually increasing.
Foreign ores of good quality, and within a reasonable distance,
are by no means unlimited in quantity. Spain, although able at
present to supply demands which are continually increasing,
shows unmistakable signs that there is a limit to her resources.
Scandinavia appears to have a more trustworthy reserve of high-
class ores, but in spite of this advantage we should long before
this have been in a very bad way had not the Thomas-Gilchrist
process rendered it possible to make good steel from second-grade
. ores. It is for this reason alone that the important deposits of
minette ores in French Lorraine, Luxemburg and Germany can
now be counted as a valuable reserve.
Mr. Martin deals also with the rate of increase in the demand
for pig iron. He recalls a prophecy of Mr. Abram S. Hewitt in
1872 that the world's production of pig iron in 1890 would reach
28,000,000 tons. The actual production for that year was 27,-
Metallurgical Notes and Comments 75.
630,000 tons. At a similar rate of increase, in seventeen years'
time the annual production of pig iron will be 80,000,000 tons,
requiring 160,000,000 tons of iron ore, that is to say, about
half as much again as the total production of iron ore in the
Bilbao district for the past twenty -seven years.
What is the lesson conveyed by these figures? The old
condition of close proximity of ore, fuel and flux has passed away.
Even in the United States the proximity of one only of these
requirements governs the location of steel works, and the diffi-
culties arising out of the transport of the others are overcome by
every conceivable device for minimizing the cost of freight and
handling. In Great Britain we are already drawing supplies of
ore not only from Spain and Scandinavia, but from Cuba, India,
South America and New Caledonia, these latter, of course, being
special varieties of ore for the manufacture of manganese and
nickel steels. But this importation from distant countries is
only now beginning. If we are going to continue to supply our
growing markets in the future, there will not be an iron-ore
deposit of good quality in any of our most distant colonies which
will not be drawn upon for British blast furnaces.
Of course a check may be put upon the expansion of our
markets. Some vSuch action is said to be about to be taken in
New South Wales. It is said that the government of that colony,
in giving out railway contracts, is specifying that nine tenths of
the material is to be of local production and manufacture. Should
this course be adopted it might naturally be expected to result
in the establishment of colonial steel works, and a corresponding
restriction in British exports.
We imagine, however, that there are more difficulties in
the wa}^ of establishing colonial steel works upon a commer-
cial basis than can be overcome by so simple an expedient as a
government bounty; for this is what the above-mentioned
restriction would amount to. For very many years attempts
have been made in India to manufacture steel and iron on
a large scale, but progress in this direction is slow. It is
probable, indeed, that the government of New South Wales
will find it necessary to begin operations on a much more modest
scale, and to build up gradually any local iron industry for which
the country may possess natural facilities. Whatever, there-
fore, may be the ultimate condition of the colonial markets for
76
The Iron and Steel Magazine
iron and steel manufactures, the coming phase promises to be
marked by a growing demand for foreign and colonial ores for
consumption in the blast furnaces of Great Britain. " The Col-
Hery Guardian," April 28, 1905.
Limitation of Chemical Analysis. — The limitation of chem-
ical analysis is well emphasized by some results recorded by Dr.
Hans Goldschmidt, with respect to the effect of titanium on the
strength of steel. In spite of practically the same chemical
composition there is a distinct difference in the physical proper-
ties of the two samples when treated with titanium thermit or not.
It is thus clear that to understand the subject, chemical analysis
must be supplemented by microscopic investigation of the
structure, and the results must be considered in the light of the
theory of soHd solutions, on the basis of Gibbs' and Roozeboom's
views. However, the effect of the titanium-thermit treatment
just noticed emphasizes another point. Here we have not simply
the production of titanium-steel. (As a matter of fact we have
yet not found any exact and clear-cut statement of the effect of
the introduction of titanium into steel.) Dr. Goldschmidt care-
fully emphasizes that at least part of the effect of the titanium-
thermit treatment is due to the mechanical stirring of the bath
and the combination of nitrogen with titanium. For exactly
the analogous reason, copper-sihcon has long been used for
copper castings, in order to combine with sihcon the oxygen of
the cuprous oxide dissolved in the copper. There is still another
point. We know that the " occlusion " of gases may have an
enormous effect on the physical properties of a metal; thus the
extreme hardness of electrolytic iron was found by Prof. C. F.
Burgess to be due to the occlusion of hydrogen in it. " Oc-
cluded " gases may play a similar important part in other metals.
However, enough has already been said to indicate the great
compUcation of the problem while the necessity of solving it is
manifest. " Electro-chemical and Metallurgical Industry,"
June, 1905.
On the Utilization of Fine Ore, Flue-Dust, Downcomer-Dust
and Stove-Dust, in the Blast Furnace. — At large furnace plants
there accumulates vast quantities of these fine ores,— flue-dust,
downcomer-dust and stove-dust,— hereafter to be referred to as.
Metallurgical Notes and Comments 77
" fine materials." It is quite a problem and considerable expense
to get rid of them. Especially is this the case in the Pittsburg,
Youngstown and Lake districts, where the furnace btirden is
made up of from 40 to 75 per cent of Mesabi ores. Not only is
this the case where Mesabi ores are used to a large extent, but
also at other furnace plants where other ores and especially con-
centrates are used.
It has been no small problem for furnace managers to work
these line materials. Many methods have Vjeen tried to get these
fine materials in a shape to be utilized in the furnace and at the
same time have the mass porous, or at least in such a shape that
when they descend with the furnace burden they will not prevent
the passage of the blast and gases.
Briqvietting has been tried and given considerable notoriety,
especially with the New Jersey concentrates, but, so far as the
writer has been able to learn, not a ton of these have been made
the past two years. This, by the way, is not because the bri-
quettes do not work well in the furnace, but because a ton of
them has thus far cost more than a ton of pig iron has often sold
for.
There is, however, I believe, one brilliant example of the
briquetting process still in use, and that is, or was, at the Joliet
plant of the Illinois Steel Company. At this plant they were
elegantly fitted to carry on this work. The furnaces were run to
produce a certain composition slag among other things, and this
slag was manufactured into cement, and a part of the cement-
making materials were utilized to incorporate with these fine
materials, which were briquetted and dried and delivered to the
furnace as a self-fluxing ore and formed a part of the regular
furnace burden. They worked well and saved hundreds of tons
of good fine materials which otherwise would have been wasted.
These fine materials are not inferior m chemical composition
to ores from which they are derived; in fact, many of them are
of considerably better composition. Analyses of these fine
materials at the Buffalo Union Furnace Company show iron
62.50, lime 7, carbon 5 per cent. Recent analyses of these fine
materials at the Port Henry plant of the Northern Iron Company,
very kindly supplied by Mr. L. D. Fraunfelder, chemist and
assistant superintendent, show theirs to run iron 51, lime 8.25
and carbon 4 per cent. These materials show, upon analysis,
y8 The Iron and Steel Magazine
that they are far too valuable to be thrown away. They are
self-fluxing and carry a good percentage of carbon to partially
reduce them.
Ore running from 50 to 62 per cent iron, on the dock at
Bufl'alo, is worth from $3.50 to $4-50 per ton.
With a 50 per cent Mesabi mixture and the three furnaces
running lull blast there will be from 40 to 60 tons of these fine
materials every twenty -four hours. Taking the average of 50
tons at $4 00, we have $200 worth of these fine materials per day.
It is useless to put them back in the furnace in their fine state,
for they will come over at once wnth the blast if put in in small
quantities, and if large amounts are filled at a time, they choke
up the furnace, thus preventing the passage of the gases and
causing the furnace to hang and slip.
The thing necessary then for the utilization of these fine
materials is to incorporate them in something which will hold
them until they get below the zone in the furnace where they can
be carried over mechanically. Many materials have been tried
and quite a number patented for bringing this about. The
principal substances employed in the lumping process are glue,
tar, molasses, asphalt, etc.
Working along these lines, the wTiter decided upon the idea
of using carbon, or coking these fine materials with bituminous
coal. The coal in coking thoroughly incorporates the fine mate-
rials, acting as a fairly good carrier, being porous, takes the
materials down to the zone of reduction and much of it still fur-
ther to the tuyeres, carries more than enough carbon for reduc-
tion, carries no foreign materials into the furnace, and besides
it is cheap. Samples have been made carrying 12 J, 25 and 50
per cent by weight of these materials. The 25 per cent one,
the mean, shows by analysis to run:
Port Henry Buffalo
^ Per cent Per cent
Iron 14-07 16.51
Lime 2.35 2. 11
Magnesium 0.30 0.70
Alumina 2.00 2.30
Fixed carbon 61.28 65.25
Volatile matter 10.56 6.86
These tests were conducted in a clay crucible. The process
can be carried on very nicely in a bee-hive oven . A small battery
Metallurgical Notes and Comments
79
of these can be installed very reasonably and made to pay at a
plant running four or more stacks. Any sized plant can be made
to pay handsomely where the by-product ovens are used as they
are at many plants to-day, and more are installing them. This
plan is also applicable at any plant where the coke is made at or
near the furnace.
I wish to acknowledge my indebtedness to Mr. L. D. Fraun-
felder, chemist and assistant superintendent at Port Henry,
N. Y., lor his cooperation and analyses. James C. Attix,
^' Journal American Chemical Society," February, 1905.
A New Vertical Illuminator. — The accompanying illustra-
tion shows a vertical illuminator of the prism type, manufac-
tured by Messrs. R. and J. Beck, Limited,
London. The device is fitted with an
iris diaphragm beneath the prism, for
cutting off outside light, and a plate of
stops so arranged that the position of
the beam of light impinging on the
prism can be varied until parallel light
of the right angle is obtained. These
vertical illuminators are used for the
examination of metals and other opaque
objects through the microscope.
New Southern Open-Hearth Steel Plant. — The Atlanta
Steel Hoop Company, Atlanta, Ga., has recently placed contracts
for a basic open-hearth .steel plant, to contain two 35-ton basic
open-hearth furnaces, with provision for a third. This will be
the third steel plant in the South, the other two being the Ens-
ley, Ala., plant of the Tennessee Coal, Iron and Railroad Com-
pany, ten 50-ton furnaces with an auxiliary Bessemer converter
and a mixer, and the Gadsden, Ala., plant of the Alabama Steel
and Wire Company, four 50-ton furnaces. The contract for the
Atlanta plant is placed with the Wellman-Seaver-Morgan Com-
pany, of Cleveland. A contract has also been placed with the
Morgan Construction Company, Worcester, Mass., for a Morgan
continuous rod mill. The Atlanta company has been rolling
hoops and cotton ties for several years from Pittsburg steeL
Basic pig iron will be obtained from the Birmingham and Chatta-
nooga districts, with a freight charge of less than a dollar a ton.
REVIEW OF THE IRON AND STEEL MARKET
The dullness in the iron trade so far as new buying is con-
cerned has continued, and is even more pronounced than a
month ago. Production has not suffered greatly. There has
been a decline of possibly a couple of million tons in the annual
production rate of pig iron from the high point, which was
reached during May. While stocks of pig iron are in some cases
uncomfortably large, the accumulation is rather small compared
with the decrease in production, and it can hardly be said that
production ot castings and rolled forms has decreased from the
maximum by as much as 3,000,000 tons a year.
The major part, if not all, of this decrease in production is
taken care of, as regards ultimate consumption, by drawing upon
the large stocks of merchant pipe, wire products, merchant
steel bars, sheets and tin plates, which were accumulated in the
winter and spring by jobbers and consumers. Actual con-
sumption has not decreased much, if at all.
There is reason to believe, on these premises, that the lull
in mill business will be only temporary. Of course there is no
expectation that there will be a resumption of active buying
before about the first of August, such a delay being natural on
account of the time of year, but general business conditions are
perfectly sound, and with reasonably good crops there is good
ground for an expectation that the whole iron trade will open
up in the fall under favorable auspices with greater activity
and satisfactory prices.
The rail trade is alone in showing marked improvement.
Recent orders for summer delivery have been so large that the
Carnegie Steel Company has just concluded to change the
Ohio mill, at Youngstown, from sheet bars and billets to rails.
It had been the intention until this decision to roll no rails at
the Ohio works this year. Further, the southern interest has
taken"'55,ooo tons of orders for delivery next year, this indicating
both confidence in the future and a desire on the part of the
railroads ordering to be assured that they would get basic open-
80
Review of the Iron and Steel Market Si
hearth rails, the Alabama mill being the onh^ maker of open-
hearth rails on a large scale.
Pig Iron. — Since our last report there has been an obvious
drop of about 50 cents in all grades of pig iron in the north, and
of about $1.25 in southern pig. The actual drop may be larger,
as the present market is one made by transactions involving
small tonnages, representing, so far as large tonnages are con-
cerned, only the nominal asking prices of furnaces. If large
orders were offered it is simply impossible to say what prices
furnaces would make. The total turnover has been very
small, but consumers are taking deliveries fairly well on old
contracts. Thus there has been little accumulation of foundr}^
pig in western Pennsylvania and Ohio, while there has been a
large accumulation of Bessemer pig, amounting, in the Mahoning
and Shenango valleys, to perhaps 100,000 tons since May i.
The difference is that foundry pig is purchased more on long-
time contracts than is the case with Bessemer, while the cessation
of buying of outside Bessemer pig iron by the United States
Steel Corporation has also had its effect. The last Corporation
purchase was for April delivery, and no further purchases are
likely in the present alignment of affairs, although sudden
changes are not uncommon in this direction. The drop in
southern pig iron has been spectacular, as a month ago $13
Birmingham was being quoted, against $13.50 ruling for a
period of several months. There followed a drop to $12.75 ^^^
then, with scarcely any intermediate prices, the market has
dropped to $11.75, with the possibility that still lower terms
could be had on firm offers. We now quote prices as follows:
F.o.b. valley furnace: Bessemer and basic, $14.75 to $15.00;
No. 2 foundry, $i4-75 to $15.00; gray forge, $14.25. Delivered
Pittsburg: Bessemer and basic, $15.60 to $15.85 ; No. 2 foundry,
$15.60 to $15.85; gray forge, S15.10. F.o.b. Birmingham:
No. 2 foundry, $11.75 to $12.00; gray forge, $11.00 to $11.25.
Delivered Philadelphia: No. 2 X foundry, $16.50 to $16.75;
standard gray forge, $15.00 to $15.25. Delivered Chicago:
Northern No. 2 foundry, $16.00 to $16.25; malleable Bessemer,
$16.25 to $16.50. Freight Birmingham to Pittsburg, $4.35;
to Cincinnati, $2.75; to Chicago, $3.65.
Steel. — The market is extremely quiet and has hardly
declared itself; quotations being made are only on small lots.
82 The Iron and Steel Magazine
and in the absence of inquiry for large lots and extended delivery
it is impossible to say what could be done. On small lots
quotations are about as follows: Soft steel billets, 4x4 and
larger, $22.50 to S23.50, f.o.b. Pittsburg; sheet bars and small
billets, $2.00 advance.
Shapes. — The structural mills continue extremely busy
on specifications against old contracts, and cannot promise
deliveries on new business for several months. Prices are
unchanged at 1.60 cents for beams and channels 3 to 15 inch
inclusive, angles 2 x 3 to 6 x 6 inclusive, and zees.
Plates. — There is not m.uch new business in plates, but
specifications on old contracts are excellent, so that all the large
plate mills are very busy. Some small mills can make early
shipments on new business, but the large mills take no interest
in this class of trade. Prices are unchanged at 1.60 cents for
plates over 14 inches and not over 100 inches wide, tank quality,
quarter-inch and heavier, with the usual advances for other
descriptions.
Merchant Bars. — The inquiries for steel bars for second
half referred to in last report as coming from buyers who had
contracts at 1.30 cents which expire July i, have been adjusted
by various means so that the buyers can get deliveries to Octo-
ber I at the same price. Really new business is quite light, and
there has been a little more shading on the part of middlemen
who had old contracts at low^er figures, this in some cases ex-
tending to $2.00 a ton, although very small lots are only obtain-
able at the official price of 1.50 cents. Iron bars have been in
better demand, prices having receded in harmony with the
much lower figures on scrap. We now quote common iron bars
at 1.55 to 1.60 cents, Pittsburg, and 1.50 to 1.55 cents, delivered
Chicago.
Sheets. — Demand from mills is fairly good, while con-
sumption is larger than would be thus indicated, as the large
stocks accumulated in the winter and spring are being heavily
drawn upon. There is no further weakness, the official prices
of 2.40 cents on black and 3.45 cents on galvanized, No. 28
gauge, being shaded from $1.00 to $2.00 per ton for carload and
larger lots.
Scrap. — The scrap market has been very quiet, but prices
have been more settled in the past fortnight, and can be quoted
Review of the Iron and Steel Market 83
approximately as follows, for Pittsburg and nearby Jdelivery :
Heavy melting stock, $14.00 to $14.50; sheet scrap, $12.50 to
$13.50; old car wheels, $15.00 to $16.00; cast borings, $8.00
to SS.50.
STATISTICS
The Iron and Steel Industry in 1904. — The general summary
of statistics for 1903 and 1904 shown on the following page is
reproduced from Mr. James M. Swank's admirable report
presented to the members of the American Iron and Steel
Association on June 10, 1905. The following interesting com-
ments are also from Mr. Swank's general review of the Am^erican
iron trade.
" The improvement in general trade conditions has par-
ticularly affected the iron trade, so much so that for several
months the demand for iron and steel products in this
country has never been equaled. It has taxed and is still taxing
our manufacturing plants to their utmost available capacity.
This extraordinary activity is probably of more general applica-
tion to all branches of the iron trade than any similar demand
in other years. Certainly our manufacturers of pig iron, steel
rails, structural steel, plates and sheets, cars and locomotives
(including railroad shops), and general machinery and foundry
products were never more actively employed than they are
to-day. The whole country urgently wants iron and steel for a
thousand uses. Our export trade in some iron and steel branches
is also contributing to the general activity. But the greatest
demand for iron and steel comes from the railroads. The fact
is now generally recognized that our railroad managers have not
kept abreast of the country's marvelous industrial development
in the last few years. More tracks, more cars and more locomo-
tives have been needed than have been built, and also more
bridges and better terminal facilities. Some of these managers
awakened to the necessit}^ of meeting these deficiencies before
the general revival of prosperous conditions last year, but
others did not awaken to the needs of their roads until the
present year, and it is to the suddenness of this awakening
that we owe much of the existing unprecedented demand for
iron and steel."
84
SUMMARY OF STATISTICS FOR 1903 AND 1904.
Subjects— Caleadax years.
Production of Iron Ore, gross tons
Imports of Iron Ore, gross tons
Production of Bituminous Coal, gross tons
Production of Pennsylvania Anthracite, gross tons...
Production of all kinds of Coal, gross tons, ,
Shipments of Pennsylvania Anthracite, gross tons...
Imports of Coal for Consumption, gross tons
Domestic Exports of Coal, gross tons
Production of Coke, net tons
Production of Pig Iron, gross tons
Production of Spiegeleisen, Ferro-manganese, and
Ferro-phosphorus, included in Pig Iron, gross tons.
Production of Bessemer Steel, gross tons ,
Production of Open-Hearth Steel, grcss tons ,
Production of Crucible Steel, gross tons
Production of Blister and Patented Steel, gross tons..
Production of all kinds of Steel, gross tons
Production of Open-Hearth Steel Castings, gross tons.
Production of all kinds of Steel Castings, gross tons..
Production of Bessemer Steel Rails, gross tons
Production of Open-Hearth Steel Rails, gross tons...
Production of Iron Rails, grosa tons
Production of all kinds of Rails, gross tons
Production of Structural Shapes, gross tons
Production of Iron and Steel Wire Rods, gross tons.
Production of Plate and Sheet Iron and Steel, except
Nail Plate, gross tons
Production of Bar, Bolt, Hoop, Skelp, Boiled Axles,
Rolled Armor Plate, etc., gross tons
Production of all Rolled Iron and Steel, including
Nail Plate and excluding Rails, gross tens.
Production of all Rolled Iron and Steel, including
both Nail Plate and Rails, gross tons
Production of Iron and Steel Cut Nails and Cut
Spikes, kegs of 100 pounds
Production of Iron and Steel Wire Nails, kegs of
100 pounds
Production of Tinplates and Teme Plates, gross tons.
Production of Ore, Pig, and Scrap Blooms for sale,
gross tons
Imports of Iron and Steel, foreign value
Exports of Iron and Steel, home value
Miles of New Railroad built (estimated for 1904)
Tonnage of Steel Vessels built in the calendar year..
Immigrants in the year ended December 31
1908.
35,019,308
980,440
252,464,776
66,613,454
319,068,229
59,362,831
3,479,430
8,312,098
25,262,360
18,009,252
192,661
8,592,829
5,829,911
102,434
9,804
14,534,978
400,348
430,265
2,946,756
45,054
667
2,992,477
1,095,813
1,503,455
2,599,665
4,962,185
10,215,220
13,207,697
1,435,893
9,631,661
480,000
9,940
$41,256,864
$99,036,865
4,715
295,840
937,371
1904.
27,600,000
487,613
249,102,765
66,318,490
314,421,255
67,492,522
1,623,280
8,673,518
23,621,620
16,497,033
220,392
7,869,140
5,908,166
83,391
9,190
13,859,887
302,834
330,211
2,137,957
146,883
871
2,284,711
949,146
1,699,028
2,421,398
4,597,497
9,728,670
12,013,381
1,283,362
11,926,661
458,000
6,743
$21,621,970
$128,663,613
4,252
100,809
808,967
RECENT PUBLICATIONS
The Chemistry of Gas Manufacture, by W. J. Atkinson
Butterfield. Third edition, revised. Volume I, Materials and
Processes. 257 5 X yj-ii^- pages; 31 illustrations. Charles
Griffin & Co. London. 1904. — In the third edition of this
well-known work it has been decided to publish it in two volumes,
the present one dealing with materials and processes and a
second volume in active preparation which will deal with the
analysis, testing and uses of gas. Volume I is divided into seven
chapters dealing respectively with the Raw Materials of Gas
Manufacture, Coal Gas, Carbureted Water Gas, Oil Gas, Enrich-
ing by Light Oils and Final Details of the Manufacture and
Sundry Schemes for Making and Enriching Gas. There is at the
present time a steadily increasing interest on the part of engi-
neers and others in the important subject of the manufacture of
gas and in the best methods of using it, and the practical book
we have before us should prove of much interest and value.
Lehrbuchder Mechanisch Metallurgischen Technologie, Zweite
abteilung, by A. Ledebur. 400 4 X 6-in. pages; over 300
illustrations. Paper covers. Friedrich Vieweg & Sohn.
Braunschweig, Germany. 1905. Price, $4.25. — This is the
second and concluding volume of Professor Ledebur's masterly
description of machine shop and other mechanical operations
to which metals are subjected, the first volume having been
reviewed in our issue for April, 1905. This treatise will un-
doubtedly be classed among the best of German technical
publications both because of the value of the contents and the
high degree of excellence of its typography and illustrating.
Electric Smelting and Refining, by Dr. W. Borchers. Trans-
lated from the third German edition, with additions by Walter
G. McMillan. Second English edition. 562, 5^ X 8^-in. pages;
254 illustrations. Charles Griffin & Co. London. 1904. Price,
86
Recent Publications 87
S7.00. — Electric smelting is daily becoming of greater import-
ance, and engineers and metallurgists are fully aware that
hardly more than an entering wedge has been driven in this
wonderfully promising field. Dr. Borchers is one of the most
authoritative writers on the subject of electro-chemistry and
this excellent translation of his important work should be
welcomed by the increasing number of persons interested in the
electric smelting and refining of metals. The first part of the
book containing 92 pages is devoted to alkalies and alkaline
earth metals; the second, containing 90 pages, to the earth
metals; and the third, covering 362 pages, to the heavy metals^
The addenda which follows contain two valuable tables, one
showing the value of equal current volumes, as expressed in
amperes, per square decimeter, per square foot and per square
inch of electrode surface ; the other shows comparison of Centi-
grade and Fahrenheit thermometer scales.
The book is well printed and finely illustrated.
Jahrbuch jur das Eisenhilttenwesen (Year Book of the Iron.
Industry), supplementary to " Stahl und Eisen," by Otto VogeL
Third year. 465 6 X 9-in. pages; illustrated. Semi-flexible
covers. A. Bagel. Dtisseldorf, Germany. 1905. Price, la
marks. — This volume gives an account of progress in all depart-
ments of the iron and steel industry during the year 1902. It
contains indexes of authors and of matter covering 40 pages.
The subject is so divided and subdivided as to make quick
reference possible, and the original source of all articles is care-
fully given. The value of this yearly publication is not to be
denied but we think that this would be greatly enhanced if
subsequent volumes could be issued more promptly.
Jahrbuch der Elektrocheniie fiir 1903, by Heinrich DanneeL
930 6J X 9i-in. pages; 137 illustrations. Paper covers. Wil-
helm Knapp. Halle-a-wS., Germany. 1905. Price, $8.50. — This
is the tenth yearly volume of a x^tiblication of growing import-
ance. It consists of intelligent and critical abstracts of all
papers and articles of value which have appeared in 1903 in
nearly one hundred periodicals. The first part is devoted to the
theoretical side of the science, the second part to applied electro-
chemistry. It also contains an index of authors of 38 pages and
88 The Iron and Steel Magazine
an index of matter of 44 pages. It is unnecessary to insist upon
the value of such work to all students of electro-chemistry and
its applications. It could hardly be overestimated. Our only
regret is that it does not also appear in the English language.
The volume is finely printed and illustrated.
Metallurgie du Zinc, by A. Lodin, chief engineer of mines
and professor of metallurgy in the School of Mines, Paris.
810 6 X 9i-in. pages; 275 illustrations and 25 plates. Paper
covers. Vve. Ch. Dunod. Paris. 1905. Price, 35 francs. —
So voluminous a book on the metallurgy of zinc suggests ex-
haustiveness, and the name of its author is a guaranty of authori-
tative and clear treatment. The first chapters deal with the
occurrence of zinc in nattire, the following ones with the pre-
liminary treatments such as roasting and calcining, while the
last chapters, forming the bulk of the work, are devoted to the
melting and distillation of zinc. Recent attempts to do away
with the retort treatment of zinc are discussed by the author.
By piiblishing such a satisfactory treatise of the technology of
zinc metallurgy the author has contributed a most valuable
addition to the metallurgical art.
Gas Engines and Producer Gas Plants, by R. E. Mathot.
Translated from the French by Waldemar B. Kaempffert, with a
preface by Dugald Clerk. 314 6 X 9-in. pages; 152 illustrations.
The Norman W. Henley Publishing Company. New York.
1905. Price, $2.50. — Gas engines and producer gas plants are
attracting the earnest attention of an increasing number of
engineers. The progress made in recent years in the construction
and running of gas engines is a notable one and augurs well for
the future of this apparatus. The translation of Mr. Mathot's
able book should therefore prove of value to many. Mr. Dugald
Clerk in his preface highly commends the author's treatment of
his subject. " I know of no work," Mr. Clerk writes, " which
has gone so fully into the details of gas-engine installation and
up-keep. The work clearly points out all the matters which
have to be attended to in getting the best work from any gas
engine under the varying circumstances of different installations
and conditions. In my view the book is a most useful one,
"w^hich deserves, and no doubt will obtain, a wide public recogni-
Recent Publications 89
tion." The book is printed in large type, on good paper and is
well illustrated.
Properties of Steel Sections, by John C. Sample. 121 6 X
9-in. pages; illustrated. McGraw Publishing Company. New-
York. 1905. Price, $3.00. — This is a reference book for
structural engineers and architects, including tables of moments
of inertia and radii of gyration of built sections, examples of
sections selected from monumental structures, unit stresses,
safe loads for columns, plate-girder designs, design in timber,
etc., with only sufficient text to explain their application. The
useful character of the book will be readily appreciated, the
aim of the author being to assist the designer in submitting on
short notice, by selection from his tables, designs which would
otherwise call for laborious calculations.
Metal Working, by Paul N. Hasluck. 760 6^ X 9-in. pages;
2,200 illustrations. Cassell & Co. London. Price, 9 shillings.
— The character of this book is thus described in the preface :
" The scope of this book embraces practically the whole art of
working metals with hand-tools and with such simple machine-
tools as the small engineering shop usually contains. The tool
outfit of the average metal worker does not generally include any-
thing more ambitious than a lathe with or without slide-rest, over-
head motion, etc., and it is with this limitation in mind that the
whole of the contents of this book have been prepared. Even
within such limits, the scope is extensive, and has been made to
include a large and pleasing variety of work."
Constitution of Hydraulic Mortars, by Henri Le Chatelier.
Translated from the French by Joseph Lathrop Mack. 132
6 X 8-in. pages; illustrated. McGraw Publishing Company.
New York. 1905. Price, $2.00. — This book is the translation
of a thesis for the doctor's degree presented by Mr. Le Chatelier
some fifteen years ago but brought up to date by a number of
notes describing the later advance in our theoretical knowledge of
cements. As the author rightly says, his thesis has been the
starting point of numerous studies, and although written com-
paratively long ago, it retains much importance for those in-
terested in cement and mortars.
■po The Iron and Steel Magazine
This late English translation should, and undoubtedly will,
receive a warm welcome. The translator justifies it in the
following words: " His classic work (Le Chatelier's) therefore
stands to-day as the first, the most complete and beautiful piece
of work done upon the chemistry of Portland cement, and since
the original is not easily obtained and all later work on this
subject goes back to and rests upon this thesis, I have thought
that it ought to be available to all who are interested in the
manufacture and use of Portland cement." The book is well
printed and attractively bound.
Cement and Concrete, by Louis Carlton Sabin. 507 6 X 9-in.
pages. McGraw Publishing Company. New York. 1905.
Price, $5.00. — This work is divided into four parts as follows:
Part I, Cement: Classification and Manufacture; Part II,
Properties of Cement and Methods of Testing; Part III, The
Preparation and Properties of Mortar and Concrete; Part IV,
Use of Mortar and Concrete. The author describes in his book
the most advanced methods of using cement in construction, and
the original investigations forming the basis of the work were
made in connection with the construction of the Poe Lock at
St. Mary's Falls Canal, Michigan. The book is well printed on
good paper, and attractively and substantially bound.
The Modern Asphalt Pavement, by Clifford Richardson,
•director New York Testing Laboratory, Long Island City, N. Y.
580 6 X 9-in. pages; 32 illustrations. John Wiley & Sons.
New York. 1905. Price, $3.00. -^ The author states in his
introduction that the object of this book is to demonstrate the
nature of asphalt pavements and the causes of defects in them,
to bring about improvement in the methods of their construction,
and to show how this can be done. The forms of construction
which have been shown by experience to be the most satisfactory
are fully described, as well as the character of the material
entering into the composition of asphalt pavements, the most
refined methods used in the industry at the present day and the
reasons which have led to their adoption, in order that engineers
and others who are responsible for the supervision and character
of such work may be able to distinguish between that of good
and that of inferior quality. To this is added specifications
Recent Publications 91
for asphalt pavement to meet various environments and uses,
and something as to their maintenance and the causes of their
deterioration. The book is well printed and bound.
The Manufacture of Hydraulic Cements, b}^ Albert Victor
Bleininger. 391 6h X 9-in. pages; 81 illustrations. Geological
Survev of Ohio. Fourth series, Bulletin No. 3. Columbus,
Ohio. 1Q04. Price, 50 cents. — In his announcement Mr.
Edward Orton, state geologist, writes that it has been his intent
and desire in authorizing and supervising the preparation of the
volume to place within the reach of every intelligent person
who desires it, a statement of what materials are needed to make
good cements, what mechanical treatment is necessary, and
what effects will result from deviations from the prescribed
quality of materials or treatment. In his letter of transmittal
he states that this bulletin represents almost four years of
unflagging and enthusiastic labor, and he expresses the belief that
it is a contribution to knowledge the importance of which will be
promptly recognized. We shall only venture a minor criticism,
namely, regarding the repeated and jarring reference to Le
Chatelier as Chatelier. The prefix " Le " before French names is
just as much a part of the name as the prefix " Mac " before some
English name, and who would think of calling Mr. Mcintosh
Intosh. The lack of culture it implies verges on offensiveness.
Electrician's Handy Book, by T. O'Conor Sloane. 761
4 J X 6-in. pages; 556 illustrations. Flexible covers; gilt
edged. The Norman W. Henley Publishing Company. New
York. 1905. Price, $3.50. — From the publishers' announce-
ment we extract the following information concerning the scope
of this book:
"It is designed to cover the field of practical electric
engineering, yet to include nothing unnecessary for the every-
day worker in electricity to know. Its pages are not encumbered
with any useless theory, everything in it is to the point, and
can be readily understood by the non -technical man, and at
the same time the educated engineer will receive great benefit
from its perusal. The Electrician's Handy Book, with its
resume of the field of electric engineering, and its logical presen-
92 The Iron and Steel Magazine
tation of the essentials of the science is a library in itself and
supplies this need."
The magnitude of Mr. Sloane's work and its great usefulness
are not to be denied. Reference books of this nature, when well
prepared, are invaluable to workers in applied science. The
publishers also have done their part most satisfactorily in
issuing the book in a convenient size, well printed and illus-
trated and suitably bound.
Annual Report a j the Smithsonian Institution, 1903, United
States National Museum. 647 6 X 9-in. pages; numerous
illustrations. Government Printing Office. Washington. 1905.
— This last report of the National Museum of the Smithsonian
Institution contains (i) the assistant secretary's report with
appendices and (2) papers descriptive of museum buildings.
The illustrations are of the usual high degree of excellence.
Contributions to Economic Geology. United States Geological
Survey. 1904. Bulletin No. 260. 620 6 X 9-in. pages.
Government Printing Office. Washington. 1905.
PATENTS
RELATING TO THE METALLURGY OF IRON AND STEEL
UNITED STATES
787,282. Meaxs for Protecting Blast-Furnace Hearths. —
Charles E. Dinkey and Hermann A. Brassert, North Braddock, Pa. A
blast furnace having a hearth
wall provided with circular rows
of separated holes extending
horizontally and inwardly from
its outer vertically extending
face below the level of the cinder
notch, and water-cooled plates
arranged in series in said holes,
the outer vertically extending
face of the masonry wall being
exposed and accessible between
the sides of the plates, and sepa-
rated horizontal retaining-bands
between the rows, the hearth wall being free from any inclosing jacket.
787,612. Method of Shaping Metallic Ingots by Transverse
Rolling. — Leonard D. Davis, Erie, Pa. A method or process of shaping
metallic ingots, which consists in first piercing the ingot with a relatively
small opening, and smoothing and compacting the walls of the opening so
made, and then transversely rolling the ingot without axial movement of
the ingot so pierced for shaping its outer surface.
787,758. Process of Treating Products Containing Vanadium,
Molybdenum, Titanium and Tungsten. — Henri L. Herrenschmidt,
Le Genest, France. A process of treating ores or products containing
vanadium for the purpose of obtaining vanadic acid, the same being
characterized by the following operations: the refining of a vanadate-of-
soda liquor by evaporation and crystallization; the precipitation of the
vanadium contained in the purified liquor by the action of concentrated
sulphuric acid upon said liquor, previously concentrated to a syrupy
condition ; and the process of reduction by a reducing agent.
787,770. Process of Manufacturing Fire-Bricks, Crucibles,
Retorts or other Refractory Articles. — Paul Klein, Riga, Russia.
A process for manufacturing refractory articles, consisting in pulverizing
chrome ore, separating the easily fusible admixtures therefrom, mixing
the chrome ore thus obtained with pulverized fire-clay and with pure
hydrate of alumina, molding and pressing and then burning such mixture.
93
94 The Iron and Steel Magazine
787,926. Process of Treating Iron, Cast Iron and Steel. —
Jean Lecarme, Paris, France. The process of hardening and transforming
into steel of variable qualities the entire or part of the surface of objects,
rotary cutters, screw-taps and other tools, and bells and other sonorous
instruments made of iron, cast iron or mild steel, said process consisting of
coating the objects on the surfaces or parts to be treated with a com-
position containing charcoal in powder, cyanide of potassium and a
combustible agglutinant body, and then heating these bodies to a bright
red excluded from the air.
788,767. Process of Coloring Steel or Iron Plate. — Thomas
O'Brien and WilHam P. Long, Elwood, Ind., assignors to American Sheet
& Tin Plate Company, Pittsburg, Pa. That method of coloring metal
sheets which consists in heating the same in a closed receptacle, reducing
the temperature of the receptacle and sheets, and thereafter introducing
an oxidizing liquid into the receptacle while the plates are hot and the
cover remains over the sheets.
788,334. Apparatus for Casting Metal. — James Scott, Pitts-
burg, Pa., assignor to American Casting Machine Company, Pittsburg,
Pa. In casting apparatus, the combination with two or more adjacent
sets of molds, having mechanism for moving them, of an intermediate
casting-pot provided with a well, and troughs integral with and leading
from it, and means for tilting the pot, whereby the liquid metal may be
directed into either set of molds, or in equal or varying streams into both
sets.
788,339. Ingot-Stripping Apparatus. — Clarence L. Taylor,
Alliance, Ohio, assignor to the Morgan Engineering Company, AlHance,
Ohio. In an ingot-stripper, the combination with a hollow rack-bar
carrying tongs, and a screw within the bar, of a plunger also within the
bar and carrying a nut which engages the screw and means for rotating
the screw.
788,378. Gas-Producer. — Josef Reuleaux, Wilkinsburg, Pa.,
assignor to Alexander Laughlin, Sewickley, Pa. A gas-producer having
an air-distributor located centrally of the fire-bed, an air-heating chamber
surrounding the producer, and a series of lateral branches connecting the
chamber and the distributor, said chamber and branches being constructed
so as to contain a body of water.
788,650. Continuous Process of Manufacturing Steel. —
Henry Knoth, Birmingham, Ala.
788,778. Process of Case-Hardening. — Carlo Lamargese, Rome,
Italy.
788,813. Process of Treating Fine Ores. — David Baker and
Wilham W. Heame, Wayne, Pa.
788,888. Apparatus for use in Uniting Iron and Stkel Plates.
— William Cross, Winnipeg, Canada.
788,964. Casting Apparatus. — Edward A. Uehling, Passaic, N. J.
789,133. Apparatus for Distributing Molten Slag in Blast
Furnaces. — Ralph Baggaley, Pittsburg, Pa., and Charles M. Allen,
Lolo, Mont.; said Allen assignor to said Baggaley.
Patents Relating to Metallurgy * 95
788,934. Manufacture of Expanded Metal. — George G. Mc-
Kay, Youngstown, Ohio, assignor to Youngstown Iron and Steel Roofing
Company, Youngstown, Ohio.
789,135. Apparatus for Charging Furnaces. — Ralph Baggaley,
Pittsburg, Pa.
789,160. Apparatus for Feeding and Distributing Molten
Material in Blast Furnaces. — Edward W. Lindquist, Chicago, 111 ,
assignor to Ralph Baggaley, Pittsburg, Pa.
789,298. Rolling-Mill Feed Mechanism. — Edwin E. Slick,
Pittsburg, Pa.
789,767. Apparatus for Charging Plates into Furnaces. - —
George T. Snyder, McKeesport, Pa., assignor to National Ttibe Company,
Pittsburg, Pa.
789,828. Ingot-Mold. — Thomas D. West, Sharpsville, and George
H. Boyd, Sharon, Pa.
789,844. Art of Controlling Furnace-Gases. — John W.
Dougherty, Steelton, Pa.
790,202. Method of Manufacturing Castings. — Jacob K.
Griffith, Latrobe, Pa.
790.269. Furnace-Wall Construction. — David Baker, Newton,
Mass.
790.270. Casting-Machine. — David Baker, Newton, Mass.
790.271. Charging Mechanism for Blast Furnaces. — David
Baker, Newton, Mass.
790.392. Process of Producing Ferro-Chromium. — Edgar F.
Price, Niagara Falls, N. Y.
790.393. Process of Smelting Iron Ores and Producing Ferro-
Chromium. — Edgar F. Price, Niagara Falls, N. Y.
790.395. Process of Producing Low-Carbon Metals or Alloys,
— Edgar F. Price, Niagara Falls, N. Y.
790.396. Process of Producing Low-Carbon Metals or Alloys.
— Edgar F. Price, Niagara Falls, N. Y.
790.397. Process of Producing Low-Carbon Metals or Alloys.
— Edgar F. Price, Niagara Falls, N. Y.
790,435. Process of Compressing Metal Ingots. — Robert W.
Hunt, Chicago, 111.
790.544. Casting Apparatus. — William S. Weston, Chicago, 111.
790.545. Casting Apparatus. — William S. Weston, Chicago, 111.
GREAT BRITAIN
5,648 of 1904. Making Slag Wool. — J. H. W. Stringfellow,
London. Improved mechanical devices for producing slag wool and
fertilizers from blast-furnace slag.
19,053 of 1904. Basic Converter. — O. Massenez, Wiesbaden,
Germany. Modification in the basic Bessemer process, whereby the
silicon is first eliminated and the slag poured off and afterward the
phosphorus eliminated.
96 ' The Iron and Steel Magazine
9,110 of 1904. Open-Hearth Furnace. — B. Talbot, Leeds. In
open-hearth furnaces, making several hearths one after the other so that
the heat and oxides contained in the slag on the finished metal on one
hearth is utilized for refining the metal in the next by causing it to boil
over the separating bridge.
9,482 of 1904. Refining Iron. — J. B. Nau, New York, U. S. A.
In refining iron by the admixture of ores, methods for keeping the ores
in thorough contact with the iron and not merely floating on top.
7,478 of 1904. Binder for Iron Ore Briquettes. — T. Rouse and
H. Cohn, London. Use of alum and water glass as binding mixture for
making briquettes of fine iron ores.
PAUL LOUIS TOUSSAINT HEROULT
SEE PAGE 165
The Iron and Steel Magazine
Vol. X
" Je veux au mond publier
d'une plume de fer sur un papier d'acier."
August, 1905
No. 2
SOME CAUSES OF FAILURE OF RAILS IN SERVICE *
ROBERT JOB, Chemist, Philadelphia & Reading Railway
'CpOR a number of years a careful study of causes of rail failures
"■- has been made by the Philadelphia & Reading Railway.
The method followed has
been, in, the case of each fail-
ure, to forward a portion of
the rail showing the defective
condition to the Test De-
partment, with a form giving
information regarding the
manufacture and service of
the rail. An investigation
then followed to determine
the cause of the failure.
In a general way it may
be said that when failure
occurs owing [to fracture, ex-
cessively^^ rapid rate of wear
under given conditions, or
crushing down in track, the
poor service is generally due to one or more of the following
causes, viz. : (i) Pipes in the steel, (2) presence of a considerable
proportion of blowholes, (3) excessive segregation, (4) coarse
granular structure, (5) rough handling.
* Read at the July, 1905, meetinj^ of the American Society for Test-
in jj Materials.
gS The Iron and Steel Magazine
In the first case the defect is readily shown by the appear-
ance of the fractured end with its unwelded surfaces, and, upon
buffing off a section of the rail, the extent of the pipe is generally
clearly shown, even without etching, as in Fig. i. In such
case the steel is in a seamy condition, and the layers readily
split apart or crush down upon comparatively slight pressure,
and indication of the unsoundness is usually given after a very
short service. This general type is caused by failure at the
mill to crop ingots down to sound steel, and such a rail is nearly
always derived from metal from the top of an ingot, and almost
invariably fails when tested under the drop test. When the
Fig. I. Pipe in Steel. Section polished but not etched
mill inspection is watched closely and when it is seen that the
test butt for the drop test is always taken from the top of the
ingot, failures due to such pipes are relatively rare.
A defect of this character is one of the most dangerous
which can be present, since it is liable to result in sudden crush-
ing under light pressure, and consequently it should be guarded
against most carefully at the inspection.
The second cause of failure, viz., presence of considerable
proportion of blowholes in the steel, is probably the most com-
mon defect under present mill practice, and the one which
causes the largest number of failures in service to-day.
Some L\jiiscs oj Failure of Rails in Service 99
Such rails usually do not fracture after a very short service
unless the extent of the blowholes is very pronounced, but the
defects are generally noticed by the gradual mashing down of
portions of the rail, accompanied generally by flowing over at
the sides of the head, and the track men are apt to complain
that the rail in question is " too soft," or that it has numerous
^' soft spots."
Analysis proves, however, that the metal is not softer than
that in other rails adjoining the defective one, but upon polish-
ing and etching the section lightly with iodine or other medium
it will be found that the steel is unsound, or, in other words,
Fig. 2. Wing Rail. Unsound steel; five weeks' service
that blowholes, slag, and other foreign matter have prevented
thorough welding of the steel and have resulted in a number
of seams which break up the solidity of the metal and permit
a slipping apart of the unwelded surfaces under moderate pres-
sure, causing final fracture or crushing. In some cases this
condition is caused by presence of slag and of oxides in the
steel, Vjut in the greater number of instances it is simply due to
blowholes, and relatively little slag or oxide is present.
Fig. 2 represents the heel of a wing rail of a frog of the
following analysis:
lOO
The Iron and Steel Magazine
Carbon 63 %
Phosphorus .137
Stilphur 078
Manganese 874
This rail crushed down in service upon another road into
the condition shown in less than five weeks.
Figs. 3 and 4 represent a batch of rails which mashed
down in a few months upon another road to the contour repre-
sented by the dotted line. The unbroken line represents the
template, and the large black areas indicate holes in the steel
which were formed by the elongation of the metal upon the top
of the rail when the failure began.
Fig. 3. Contour of Fig. 4 after service of five months
When this section was polished there was no indication of
an actual pipe, but upon light etching the cause of the weakness
is very evident, showing that the steel is so thoroughly unsound
and porous that rapid crushing was possible.
The composition of this rail was:
Carbon 82 %
! Phosphorus 102
Manganese 88
Sulphur 053
and the heat average was :
Some Causes of Failure of Rails in Service loi
/O
Carbon 59
Phosphcn'us 075
Sulphur 074
Manganese 1.06
It is thus seen that a marked segregation was present, as
well as great unsoundness, and it is evident that the cropping
at the mill had been insufficient to get to sound and reasonably
homogeneous metal. These rails were furnished under the
mill's own specifications and guaranty.
The tendency of some rails to flow over and form a '' lip "
Fig. 4. Unsound Steel; five months' service
has been referred to, and this frequently is attributed to mere
softness of the steel. Our experience has been that a rail with
carbon as low as even 0.33 per cent will not flow over under
exceedingly heavy traffic, provided sound steel is present, with
granular form fine enough to render the metal tough and strong ;
and in every instance of flowing over or of breaking down of the
side or corner of the head, we have found presence of blowholes
or other unsoundness near the surface or corners of the head of
the rail, generally within one eighth or one quarter of an inch
of the surface, whereas in the cases in which the rails have
sustained long and heavy traffic, we have found comparative
I02
The Iron and Steel Magazine
freedom from such defects. A case of this kind which was
investigated a couple of years ago will illustrate the point.
A number of rails from a single rolling had mashed down in
track after two months' wear, to the general form represented
by Fig. 5, the broken line representing the template. These
defective rails were upon curves and upon tangent, and were
from various heats throughout the rolling; but rails from the
same heats and immediately adjoining the defective rails were
unaffected by the traffic.
It was found that composition had nothing to do with the
failures, but upon polishing and etching the sections we found,
in the cases in which flowing and splitting occurred, that
'
f^ "^ — "
Vi
"---^ ■
'
^^^ r'^
Fig. 5. Contotir of Fig. 6; two months' service
the steel was unsound, as indicated in Fig. 6, while the rails
adjoining, which were in good condition, were of sound steel, as
indicated in Fig. 7.
We have also investigated a considerable number of rails
to determine whether or not the form of section exerted an
influence upon the tendency to crush in service, and it has been
clearly proved that ability to withstand crushing under heavy
service is due not so much to any particular form of section, as
to the relative freedom from unsoundness in the metal. In
connection with this question an examination was made about
a year ago upon a lot of rails which were removed from main
track and relaid in branch lines. The section had a rather deep
Some Causes of Failure of Rails in Service 103
head, with contour approximating that of a wheel flange. Prac-
tically no flowing and no crushing was found in these rails after
fifteen years' seryice, while a number of rails of the American
Society section laid in adjoining parts of the same track had
crushed and otherwise failed in a very few years. The compo-
sition of the different lots showed comparatively little variation,
but in the case of the rails which crushed in service we invariably
found unsoundness, of the general type indicated by Fig. 6,
whereas in the good rails the steel was practically sound within
one-half inch of the surface of the top and sides of the head. In
Fig. 6. Unsound Steel; two months' service.
Mashed down as shown in Fig. 5
other words, the good service was due mainly to the greater
care exercised in the manufacture of the earlier rails and their
consequent relative freedom from unsoundness.
Up to this point we have made little mention of granular
structure of the metal, and it may be inferred that this has little
influence upon the permanency of the rail in track. On the
contrary, we have reason to regard a uniform fine granular
structure of high importance, both in reducing rate of wear and
in cutting down liability of fracture, but it is an unfortunate
fact that the very best of steel as regards granular form or com-
I04 The Iron and Steel Magazine
position may be cdffipietely and quickly ruined, from the stand-
point of efficiency in track, or, in other words, as to its value
as a rail, by failure in manufacture to insure reasonable freedom
from unsoundness. This, to our mind, is an integral point in
the manufacture of rails, for if any material degree of unsound-
ness exists in the rail within a distance, say, of one quarter or
one half an inch from the head or sides, and more particularly
near the upper corners, unsatisfactory service under heavy
traffic is almost certain to result, regardless of the composition
Fig. 7. Sound Steel ; two months' service. Rail next in track
to Fig. 5. No tendency to mash down or flow over
or method of rolling. Such, at least, has been our invariable
experience.
As to injurious segregation, we find that relatively few
failures in track are due to this condition. If ingots are not
properly cropped, or if they are allowed to remain in the furnace
with the interior of the ingot in a liquid condition for an exces-
sive time, segregation, of course, results, and if the test butt
from each heat is taken, as should invariably be done, from the
top of the ingot, badly segregated rails will fail, and conse-
quently, under such conditions, careful guarding against segre-
gation is as much for the interest of the manufacturer as for
So)}ic Causes of Failure of Rails in Service
105
that of the consumer. Fig. 8 represents one such rail which
failed under the drop test, and the extent of the segregation is
shown by the fact that the proportion of carbon at the outside
averaged 0.49 per cent, while at the center of head borings
taken with a quarter-inch drill averaged 0.76 per cent.
Also, it will be noted that blowholes extend all along the top
of the head.
Such cases as this are, however, rare.
Under '' failures due to rough handling " we have found a
considerable number from time to time. The initial fracture
Fig. 8. Segregated Rail. Broke under drop test
may, of course, occur either at the mill at straightening presses,
or in loading into high side cars and letting fall upon other rails
five feet or more below, or in letting the rails fall from the loader
•six or seven feet upon the ground, and the same thing, of course,
may occur after receipt of the rails in unloading unless they are
skidded out or otherwise gotten to the ground without any
considerable shock. Careful inspection will remedy this con-
dition. A fracture of this type is characteristic. It begins
generally across the base of the rail and extends up a short
•distance into the web, then it works along the web sometimes
io6 The Iron and Steel Magazine
for a distance of six feet or more, with the face of the fracture
in a plane at right angles to a vertical line down through the
rail, and the steel finally snaps off up through the head.
A fracture due to this cause can generally be identified at
a glance and can be distinguished from a fracture caused by
pipes, since the latter extend with the unwelded faces more or
less parallel with the contour of the rail.
To sum up, the results of our investigation indicate that
the greater part of the difficulty which occurs to-day with rails
under heavy traffic is due to unsound condition of the steel, a
condition which existed in comparatively slight degree in the
earlier rails.
There has been a marked improvement in practice at some
of the mills over that generally prevalent a decade or two ago,
and this has resulted at such mills in producing a much finer
granular form throughout the section and hence a tougher and
better wearing rail if only the metal were sound, but unfortu-
nately, in the essential element of soundness of the steel there
has been direct retrogression, making it appear that the main
attention in the manufacture has been fixed upon quantity
and not quality of the output.
Reading, Pa., June 29, 1905.
Standard Methods for Testing Cast Iron
107
A COMPARISON OF STANDARD METHODS FOR TESTING
CAST IRON*
By DR. RICHARD MOLDENKE
TX reviewing the situation as it exists to-day, we see that all
the work carried through in connection with the testing
of cast iron, lies in the
direction of standard'speci-
fications. The only ^na-
tions which have accom-
plished something definite
are Germany and the
United States. The others
are still working at the
problem.
As pig iron is the
basis of the foundry in-
dustry, our attention is
first directed to it, and we
find two general specifica-
tions in use: the Ameri-
can ones, and lately a pig
iron contract drawn up in England. In the American specifica-
tions we have the direct recommendation that all iron be pur-
chased by analysis. Next, detailed instructions for sampling, —
the course to pursue in case of a disagreement in the analytical
work. An important omission, and one which it will take much
time to supply, is the adoption of standard methods for analysis.
Without these, even the best specifications still leave a loop-hole
for controversy. Incidentally, it may be said that the American
organization of foundry men is taking this matter up, and has
already prepared a standard method for determining silicon in
pig iron and cast iron. Total carbon is to follow, then sulphur,
and so on. As these methods are tested out in practice outside
of the foundrymen's organization, we shall learn their practical
value for specification purposes better.
Continuing with the American pig-iron specifications, we
next come to the allowances and penalties. Here there is given
* July, 1905, meeting of the American Society for Testing Materials.
io8 The Iron and Steel Magazine
the limit of difference allowable in the pig iron delivered from
that specified, and the penalty that may be exacted where the
limit is exceeded, and yet not be too great to absolutely reject
the metal. These provisions enable foundries to purchase pig
iron with a reasonable assurance that they get what they want
nearly enough, without causing the slightest trouble in the shop
routine. The cash penalty further prevents the furnace from
taking chances on shipments to people who watch their supplies
carefully.
For the benefit of the trade in general, inasmuch as only
the minority of foundries are equipped with laboratories, or
have expert advice, there is given a table of base analyses of
grades, so that if a man pins the specifications to his order, and
calls for a No. 3 iron, he will get just what a. No. 3 iron should
be in composition, so far as the silicon and sulphur is con-
cerned. When the use of chemistry in the foundry is so
general, and the furnaces are run in such a way that one iron is
as good as another, we may see these specifications extended to
include the other elements. At present the phosphorus, man-
ganese and carbons are questions of brand and locality largely,
the furnace industry being quite settled in classes for pig-iron
distribution.
Germany has not yet seen fit to standardize pig irons, and
reports from the other side indicate that conditions are not so
favorable there, the application of our American specifications
being out of the question for German irons. A man calling for
pig iron with the sulphur we give, and be it said that our sulphur
limit is high, would have to pay fancy prices in Germany, for
they are badly troubled with that element over there. The
same may be said for England, and, on looking over the new
pig-iron contract issued by the London Metal Exchange, we
find that while the sulphur allowed is not much larger than ours,
yet the very much higher silicon that goes with it practically
makes a big difference. Thus, while we have a No. 2 pig iron
run 2.25 silicon, with a variation of 10 per cent either way, or
2.00 to 2.50, and this has a maximum of .045 sulphur allowed,
the English standard, with the same sulphur, allows the silicon to
vary from 2.50 to 3.50 per cent, which would correspond to our
No. I with a higher sulphur. The English specifications also give
rules for sampling, but lay much stress upon the brand names.
StathhirJ Methods of Testing Cast Iron 109
Incidentally it may be mentioned that America is taking
up the question of standardizing foundry coke, which is a step
in advance, and will have a far-reaching influence, not only in
foundry practice but on the blast furnace.
To turn now to specifications for testing cast iron. In Amer-
ica we have adopted a set for pipe, for locomotive cylinders,
for malleable castings, and there are pending those for castings
in general, and for car wheels. Over here we take out from the
general work, the special groups, which can stand by them-
selves and have properties peculiarly their own, which may be
determined by specific tests. In Germany they have specifi-
cations fcr machinery castings, for columns (which we are trying
to get away from as quickly as possible) and for pipe.
In dividing the classes of castings relative to their thickness,
for this is the important point to consider when specifying
breaking strength, we have adopted a little wider limits than
the Germans. Thus we have small castings at h inch and less.
They have 0.6 inches and less, or a little more. For medium
castings, however, we have from \ inch to 2 inches. The Ger-
mans have from 0.6 inch to only i.oo inch. For heavy castings,.
we have over 2 inches in thickness, and they have over i.oo
inch, which shows that our conception of heavy castings is a
little different, or else that German customs lay more stress on.
smaller limits for medium castings.
A further difference between the American and the German
specifications may be found in the chemical end. We specify
the upper limit for sulphur, so as to secure reasonable strength
against shock. This is not looked after in the German specifi-
cations, possibly because of the difficulty in getting low sulphur
irons for the foundry.
The point that interests us most, however, is the method by
which the metal is judged; that is, the test bars emploved.
Comparing the general specifications advanced for Germany
with our own, now pending, we see that special pains are taken
in both cases to get representative test bars, and these are not
to be cast on the piece. Herein there is a distinct advance,
cutting off the old coupon. The transverse test is prescribed,
which agrees with our experience. The tensile test is omitted
entirely in Germany, and it is to be hoped that we may follow
suit in this some day also, as no good end is served when no
no The Iron and Steel Magazine
two testing machines may agree in the alignment and grip on
the specimens.
We find a radical difference in the length of the test bars
used. Our own are comparatively short, and this has caused
comment on the other side, our German brethren concluding
that we do not lay as much stress on the transverse test as we
should. We, on the other hand, believe that with the long bars
in use formerly, much of the sensitiveness of the transverse test
is gone, for even poor iron will show good results, if the test is
■carried out slowly and carefully. On the other hand, with a
comparatively quick test on short bars, the iron must be of
goo4 quality to show a good deflection and strength.
Three bars are provided by the Germans : for small castings
the diameter is 0.8 inch and the distance between supports 16
inches. For the medium castings the figures are 1.2 inches
diameter and 24 inches between supports. For the heavy
castings the diameter is 1.6 inches and the testing distance 32
inches.
It will be noted that the German aim is evidently to get as
near the size of the castings to be represented as possible, and
this is to be commended in a way. However, we realize over
here that the lack of homogeneity in the structure of cast iron
is such an element in the problem that the records of several
sized test bars are not mathematically comparable, as would
be the case in steel. Hence we would not feel safe to accept
the result of a long and thick bar as compared with a shorter
and thinner one, in order to judge whether the iron in one is
better than that in the other.
While realizing that it is desirable to vary the diameter of
the bars, but not the length, we reluctantly confined ourselves
to one bar for all purposes, aiming only to get at the actual
quality of the metal with given standard conditions, identical
for each test, so far as foundry practice can accomplish this.
We can, therefore, discriminate between metal wanted for light,
medium and heavy castings at a glance, and without making a
comparative calculation, the results of which are open to doubt.
The German specifications for casting the test bars go us
one better in requiring the vertical pour, but from bottom up.
We await their results on this with interest, as we use the ordi-
nary top pour, but so arranged that the metal drops to the bot-
StajiddrJ Methods of Tcstiiiii (\ist Iron i:i
toni through a funnel-shaped gate, and the mold is thus made
cheaper.
German specifications call for the bars to be made in flasks
that are not parted, if possible, so that the test bar has
seams. If, however, this is unavoidable, the test bar is to be so
placed that when tested the seam lies in the neutral axis. We
prefer to prevent test bars made with seams altogether by giving
complete specifications for the flask itself, which any foundry
can arrange for without particular trouble.
Both specifications agree in having the bars cast in dry
sand, and the cooling of the bars in the flask. Furthermore,
only brushing is allowed in cleaning the bars, and no machining'
is to be done.
In judging the tests themselves there is a difference between
the two specifications in question. We specify just when the
tests are to be arranged for in the heat, and that one of the two
bars cast at the various casting intervals must pass the require-
ments. The German specifications call for three bars, the
average of which must be taken, defective bars to be excluded.
In both cases the expense of testing falls upon the founder. In
our added tensile test, this, when required by the purchaser, is
to be paid for by him.
The clause in our specifications wherein we allow the buyer
the freest run of our establishments, in order that he may be
satisfied that the material is gotten out in the best manner
possible, does not appear in the German specifications. Only
in the case of pipe is there mention made of facilities to be given
the inspector to watch the testing of this material.
It is still a little early to draw conclusions from the specifi-
cations advanced, for they have either not been officially adopted
by their respective countries, or they are still in the trial stage.
This much can be said, however, that a marked advance can be re-
corded, for in everything presented so far, the attempt has been
made to build on our increasing knowledge of the properties of
cast iron as a metal. Much has, of course, to be yielded to busi-
ness expediency, for the industrial customs of a nation cannot
be radically disturbed without laying ourselves open to the charge
of being idealists and dreamers.
The buying of pig iron by analysis, and now by specifica-
tion, may be said to be the most radical advance the foundry
112
The Iron and Steel Mamzine
a^
has ever made. The adoption of specification for castings is
gradually coming into vogue also, and we will soon see the allied
industries, such as the fuel, sand, facings, etc., become a subject
for study and final specification.
It is to be hoped that at Brussels, next year, we not only
may report final specifications for all we have undertaken in the
way of cast iron, but that Germany, England, France and
Austria may be similarly situated. Then we can compare notes,
and possibly adjust some of the items so as to have a greater
conformity in practice.
RECENT DEVELOPMENTS OF THE BERTRAND-THIEL
PROCESS IN THE MANUFACTURE OF STEEL *
By JOHN H. DARBY, Brymbo, and GEORGE HATTON, Round Oak
"ly yTANY attempts have been made to accelerate the open-
^ hearth process of steel making, acid or basic, by increas-
ing the size of the furnaces employed, perfecting the machinery
for charging the steel-making materials, using molten iron
from the blast furnaces, regulating its supply and composition
by the use of a mixer, etc., and the object in view has more or
less been accomplished. But with all such improvements, in
practice it is generally found essential to employ pig metal of
special composition, which is not always readily obtained, and
may be somewhat costly to manufacture.
The authors hope to show that by the aid of the Bertrand-
Thiel process a variety of pig irons may be used, differing in
composition, while at the same time the output of steel is kept
at its maximum, and the quality of the product uniformly ex-
cellent. A. Ledebur t observes '' that if two substances react
on each other chemically, this reaction proceeds more slowly
the more the two substances are diluted by other substances
which remain inert. In other words, the greater the excess
of one of the two reacting substances the more quickly is the
chemical conversion of the other completed." This, in the
authors' opinion, is the basis for the use of two furnaces in
the Bertrand-Thiel process.
* Iron and Steel Institute, May, 1905, meeting. Slightly abridged,
t " Stahl nnd Eisen," Vol. XXIII, pp. 36-41.
Bcrtraiid-Thicl Process in the Ahuiujacture of Steel 113
When in 1894 Messrs. Bertrand and Thiel introduced their
process at Kladno. it was the practice, as open-hearth furnaces
existed at a suitable level, to hold the molten iron in these fur-
naces ready for the basic converters, and to effect a certain
amount of refining, by additions of ore and lime, in order to
prepare the metal for blowing. A 20-ton open-hearth furnace
also existed at a lower level, sufficiently near to be commanded
by a long trough connecting the upper and lower furnaces.
With these conditions, it occurred to Bertrand and Thiel that
preliminary refining, or roughing down, might with advantage
take place in the upper or primary furnace, and that when the
bulk of the phosphorus, and incidentally the whole of the silicon
with part of the manganese and carbon, had been removed,
they would have metal highly suitable as regards composition
and temperature for transfer to a secondary furnace, in which
suitable proportions of scrap, iron ore and lime had been pre-
viously charged and heated, care being taken to prevent the
slag from the primary furnace passing to the secondary furnace
during the transfer of the metal.
Prof. H. M. Howe has aptly described the process as analo-
gous to washing one's hands in water, so making them moder-
ately clean, and subsequently changing to fresh water, where
the final degree of cleansing can be attained.
As by the medium of the first washing a moderate degree
of cleansing is attained, so the feebly basic slag in the primary
furnace is able to retain a large percentage of the phosphoric
acid and silicon liberated by the oxide of iron. Slags from
the primary furnace frequently hold double the percentage of
phosphoric acid compared with the lime base.
The following is a typical analysis of primary slag when
using phosphoric pig iron:
SiO^ 10.22%
P2O5 28.86
CaO 18.60
FcoOg 6.70
Notice how feebly basic the slag is, and yet how perfectly
the iron in the ore has been exhausted, while the slag is largely
charged with phosphoric acid. A yield of 103 to 105 of steel
ingots per 100 of metal is attained.
114 ^^^^ Iron and Steel Magazine
This means saving in lime consumed to fix the phosphorus
removed from the pig treated, and insures a high percentage
of the phosphoric acid in the slag, usually over 90 per cent of
which is soluble according to the standard test.
As Bertrand worked his process, six to seven 20-ton charges
of soft steel per twenty -four hours were considered good practice
for a pair of furnaces. At Brymbo, with a more highly phos-
phoric pig, seven similar charges per day during a week's working
have been attained. At the Hoesch works in Dortmund ten
charges have been regularly produced per day.
Fast working means a reduction in the cost of conversion, so
that at present, taking into consideration the outlay involved,
the Bertrand process may claim to be operated as economically
as any other known method of converting pig metal into steel.
Bertrand's original plant, with furnaces at different levels,
can only be used under special conditions, and had the disad-
vantage that when one furnace was under repairs the other had
to remain idle.
Many works have their furnaces arranged in line, and to
suit this construction a convenient arrangement is to place a
mixer at one end or in the center of a line of furnaces, com-
m.anding the whole by powerful overhead cranes and charging
machines to charge the metal from the mixer to the primary
furnace, and, after partial purification, to transfer again to the
secondary furnace for final treatment.
An arrangement of plant well suited for the process is shown
on Figs. I, 2 and 3 herewith.
Any scrap that it is desirable to melt up and make use of is
charged into the secondary furnace, and the quantity so used
is only limited by the time occupied in charging the furnace
and melting. A pair of furnaces, working 16 tons of phos-
phoric pig in the primary, and 20 tons with scrap in the sec-
ondary furnace, would tap at least every two and one-half hours,
so that charging by pouring in molten metal and mechanical
charging of the scrap are desirable, and even necessary, to obtain
the best results.
A mixer is no part of the Bertrand-Thiel process any more
than it is of the Bessemer or Thomas processes.
In the case of two works employing the Bertrand-Thiel pro-
cess, the pig iron containing a high percentage of phosphorus,
Bcrira)id-Thicl Process in the Mauujaciurc of Steel 115
ii6
The Iron and Steel Magazine
it was found advisable to employ a gas-heated mixer, in fact, a
large tipping gas-heated furnace. In it the molten metal from
the blast furnace is poured and retained for use as required,
week-end metal melted, desulphurization takes place, and the
percentage of silicon is reduced. If a temperature of about
1500° C. is maintained in the furnace, cold pig iron is readily
melted, and the wear and tear of the furnace is slight.
Average
Composition
of Metal
charged
to Mixer
Average
Composition
of Mixer
Metal ready
for Steel
Furnaces
Average
Composition
of Primary
Metal
Average
Composition
of Final
Metal
Carbon . .
Silicon . .
Sulphur . .
Phosphorus
Manganese
Per Cent
3.250
0.654
0.076
2.420
2.400
Per Cent
0.473
0040
0.510
Per Cent
1.800
0.080
0.036
0.580
Per Cent
0,11
0.46
0.028
0.035
Note the reduction in sulphur and silicon in the mixer,
and that the carbon remains almost untouched.
The mixer is a costly piece of apparatus, and in order to
get a better return in work done than is described above, it
was decided, as an alternative to working the process as origi-
nally employed by Bertrand, to carry the refining in the mixer
still further, only raising the temperature sufficiently high to
melt the slag while adding a considerable quantity of iron ore
and lime.
Average Composition
of Pig Metal charged
into Mixer as
Primary Furnace
Average Composition
of Refined Metal
from Mixer as
Primary Furnace
ready for
Secondary Furnaces
Average Composition
of Final Steel from
Secondary Furnace
Carlson
Silicon
Sulphur
Phosphorus ....
Manganese ....
Per Cent
3-25
0.636
0.053
2.42
2.40
Per Cent
2.10
0.160
0.043
0.930
0.78
Per Cent
0.169
0.042
0.028
0.032
The average temperature of the furnace, looking on to the
surface of slag through the central door, was 1545° C. as shown
Bcrtrand-Thicl Process in the Manufacture of Steel 117
bv the Wanner pyrometer, compared with the secondar}^ fur-
nace in full heat of 1720° C.
It will be noticed that the carbon largely remained, the
silicon was almost all eliminated, the sulphur reduced and
61.5 per cent of phosphorus removed. Counting the mixer
as a primary furnace, three furnaces were engaged, the two
secondary furnaces charged with 25 per cent scrap working up
the output of the mixer with forty-two 20-ton charges between
three furnaces, or fourteen charges per week per furnace.
The added ore in the mixer was well exhausted. The speed
of working was slower than the first modification of the process
tried.
Average composition of mixer slag:
Silica 15-5%
Ferric oxide 9.0
Phosphoric acid 18.5
Lime 19.5
The original method of working employed by Bertrand has
the advantage of faster working. The second method, although
not so expeditious, involves less trouble in transfer, but is trying
to the lining of an important furnace like the mixer.
In both the methods described phosphoric iron was em-
ployed, made with specially low percentages of silicon and sul-
phur.
As it had been noticed that sulphur was very readily elimi-
nated in the primary furnace, irons were obtained containing
high percentages of sulphur and silicon in order to test the
process.
The following are notable instances:
Percentage of
Sulphur in Pig
Metal
Percentage of
Sulphur in Primary
Metal
Duration of 1 8-Ton
Charge in Primary
Furnace
Percentage of
Carbon in Primary
Metal
Per Cent
0.308
0.209
0.240
0.370
Per Cent
0.060
0.070
0.078
O.IOO
Hours Mins.
2 45
2 15
2 40
3
Per Cent
2.2
1.6
1.8
0.9
The percentage of sulphur in the finished steel from the
secondary furnace did not in any case exceed 0.05 per cent.
II.
The Iron and Steel Magazine
Pigs with 2 per cent to 2.5 per cent of silicon are readily
treated in the primary furnace, and owing to the short duration
of the charge no excessive wear of the lining takes place.
From experience gained the authors can state that by
partial preliminary refining in one furnace, pig of almost any
ordinary composition can be commercially treated.
T
THE CONTINUOUS STEEL PROCESS IN FIXED FURNACES *
By S. SURZYCKI, Czenstochowa, Poland
'HE process is based on the idea and the principle of the
Talbot process, with the essential difference that it can be
carried out in any fixed furnace of not less than twenty -five tons
capacity. This practice is mainly based on the arrangement
of two or more tapping holes placed over each other, but not
arranged in a vertical line, such tapping holes leading into a
double launder by which the whole of the contents of the fur-
nace, or only a part of the contents, may be easily emptied at
any time. The tapping holes are arranged in a cast-iron plate,
lined up with magnesite bricks and stamped dolomite; they
give no trouble since they can neither slip nor sink, nor yet be
corroded. The closing of the tap holes is carried out as follows :
When the molten bath has so far sunk that it is on a level with
the bottom of the tap hole, it is agitated by means of a rabble,
by which means the bottom of the tap hole is perfectly freed
from metal. Immediately thereupon some shovelfuls of dry
burned dolomite are thrown against the tap hole from the charging
side of the furnace, and the last drainings of metal are removed
from the tapping side, and it is then made fast on that side.
The whole operation lasts some few minutes.
The hearth of the furnace stands perfectly, and remains
sound for months without any repairs. As a rule only the slag
line, and to a less extent the blocks, require repair.
At the Czenstochowa works there is at work a 2 5 -ton furnace
of the following dimensions: The length between blocks, 26
feet, 3 inches; width, 8 feet, 7 inches; and depth of hearth from
I foot, 8 inches, to 2 feet. During its first campaign it made
* Iron and Steel Institute, May, 1905, meeting. Slightly abridged.
The Coiiti}iuous Steel Process in Fixed Furnaces 119
574 char^cfes by the continuous process, whereupon, while still
in quite good condition, it was stopped for certain local reasons.
After repairing the ports and the roof and repacking the cham-
bers, the furnace was again put into work, and since then has
already made over 690 charges, and will, it is hoped, keep at work
up to the 1,000 charges.
The chief material required for the continuous process is,
of course, pig iron, employed in a molten condition. On this
point the conditions under which the furnace has to work have
not been specially favorable. The pig iron taken from a 70-ton
mixer, into which it is run from a single blast furnace, could be
improved both as to homogeneity and quality. This molten
pig iron is also taken from the same blast furnace for other
small furnaces at the same time. The variations in the chemical
composition of the pig iron are very considerable, as the fol-
lowing shows :
Carbon up to 3 %
Graphite up to 3.7
Silicon 0.8 to 1.9
Manganese 0.6 to 1.5
Phosphorus 0.5 to 0.8
Sulphur 0.02 to o.io
These variations have a great influence on the production
of the furnace, and in order to lessen them as much as possible
the hotter and more graphitic pig iron is treated with iron ore
(Krivoi-Rog ore with about 63 per cent of iron). This is a
well-known method in the foundry, and consists simply in throw-
ing with shovels the fine previously heated and dried ore into
the stream as it runs into the movable ladle.* There is pro-
duced thereby a more or less intense reaction, according to the
temperature and according to the composition of the pig iron,
which promotes the conversion of graphite into combined car-
bon, and the oxidation of the silicon and manganese.
The pig iron, somewhat refined by this treatment with ore,
when brought into the furnace requires less ore there, and
works down quicker. The production of the furnace is very
closely dependent on the quality of the molten pig iron. It
is clear that the graphitic pig iron rich in silicon will take much
* Mr. R. M. Daelen formerly constructed an apparatus for this
purpose.
120 The Iron and Steel Magazine
longer to work down than one without graphite and with a
less percentage of silicon; especially is the influence of large
variations in the chemical composition of the pig iron of im-
portance, as the ore and lime additions have to be measured by
the eye.
This composition plays a role in this process only in refer-
ence to the graphite and sulphur. As before stated, pig iron
rich in graphite is difficult to treat in the open-hearth furnaces,
the percentage of manganese is not essential, and is only so
far of importance that a higher percentage of manganese retards
the refining process. The percentage of phosphorus can go
very high; with more than i per cent of phosphorus the slag
must be drawn off as quickly as possible after the reaction, that
is, after the end of the refining period, since it is known that the
reduction of the phosphorus from the slag can take place, which
must be avoided; a very rich phosphoric slag, which can be
sold for agricultural purposes, can be obtained.
The furnace makes usually three charges in the twenty four
hours, each of 25 tons; when the blast furnace is working well,
and a pig iron with low silicon and manganese is obtainable,
four charges can be made. The following record of the work
of the furnace for a recent month illustrates the results obtained
by this process :
Consumed
Kilos per i Ton
Tons of Good Ingots
Cold pig iron 140.80 70.3
Molten pig iron 1,809.50 903.1
Ferro-manganese 20.9 10.4
Scrap 19-8 9-0
Iron ore (Krivoi-Rog) 458.1 228.6
Lime i44-3 72-o
Aluminum 0.08 . . .
Burned dolomite io7-7 53-8
Chrome ore 2.0 i.o
Produced
Good ingots 2,003. 7 tons
Scrap 41.8 tons
Working days 26
Production per day 77-o7 tons
Yield 102.72%
The Cojit ill nous Siccl Process in Fixed Furnaces 121
From the above figures it is seen that the yield (reckoned
on the metal charged) is 102.7 P^^ cent; but it must be stated
that the daily reports are made up with a certain amount of
circumspection, and it may be taken that the average yield is
not less than 103^ to 105 per cent. From this one could easily
reckon the reduction of the iron from the ore. In the Krivoi-
Rog ore there is an average of 63 per cent of iron; that is to say,
in the whole amount of ore introduced there are 288.6 tons of
iron. Since in the pig iron added there is 5 per cent of foreign
bodies, and for the other metals added 5 per cent of loss can be
taken, there has been, if we take the average yield at 104 per
cent, 180 tons of iron or 62.3 per cent reduced. This reduction
of iron from the ore in an essentially oxidizing apparatus, such
as the open-hearth furnace, must be looked upon as a great
advantage, and it is scarcely possible in any other refining opera-
tion, even in the Talbot furnace, to obtain a greater reduction.
Unfortunately, I am not able to give exact figures as to coal
consumption, since the producer plant is common to the whole
of the furnaces, and it will only be when the new continuous
furnace, now building, which will have separate producers, is
working that it will be possible to give the coal consumption
exactly. Of course, it is considerably less than the ordinary
process, as the furnace when continuously worked stores up an
immense amount of heat, and much less fresh gas than in the
ordinarily worked furnaces is required. It can, however, be
stated that the coal consumption for the whole steel plant (which
consists of five furnaces, four of which are ordinarily in opera-
tion), since the introduction of the continuous process, has
fallen about 9 or 10 per cent (reckoned on the production of
good ingots).
A few words on the quality of the material produced may
be useful, as this subject does not appear to have been very
clearly understood.
The making ready of the charge, according to the method
described above, and also according to the Talbot process in
tipping furnaces, takes place partly in the steel ladle. I say
advisedly " partly," as actually the charge must be finished
sufficiently in the furnace itself. The test taken from the fur-
nace before tapping, and before the addition of ferro-manganese,
must be capable of being hammered out without any fault,
12 2 The Iron and Steel Mamzine
is'-
must possess the wished-for hardness, must be perfectly dephos-
phorized, must show a perfectly sound fracture, and must remain
quiet in the small test ingot mold. This being so, the charge
can then be tapped, and at the same moment a regulated amount
of finely ground ferro-manganese is thrown against the stream
of steel as it falls into the ladle. A fairly active reaction is there-
by produced, which after one to one and a half minutes quiets
down completely, and when the metal has stood for a short time
teeming can be commenced. In order to treat the charge
rightly by this method undoubtedly a certain amount of ex-
perience must be had; but it appears incomprehensible why by
this method just as good a quality of steel should not be obtained,
as some steel works managers have asserted. In the first place,
the quality of the final product is dependent on the quality of
the pig iron employed. It is clear that from a pig iron rich in
sulphur or copper it is more difficult to produce a good quality
than from a good first-class pig iron. In this connection one
can always help matters by the addition of Spiegel or in other
ways even more easily in the continuous process, as the impuri-
ties in the pig iron are taken out more easily and with less trouble
by admixing it with the soft, pure bath of metal remaining in
the furnace than in the ordinary process, where one would have
to treat the bad pig iron alone. It must be observed that in
the 2 5 -ton furnace with the deeper hearth, 25 tons, as a rule,
are tapped, and the same quantity of molten pig iron poured
in, 15 tons of pure soft metal being left behind. Thus the capac-
ity of the furnace by this process is increased to 40 or 45 tons,
although it was constructed as a 2 5 -ton furnace only. In the
second place the dephosphorization by means of the slag goes
on quicker in the continuous process, since the pig iron while
being poured into the furnace is already partly dephosphorized
during the reaction which takes place. Further, the charge
must be very carefully worked down before tapping and before
the addition of the ferro-manganese, since the influence of the
calculated amount of ferro-manganese, which is added in the
ladle, is only to bring about the deoxidation of the metal and
the introduction of the necessary amount of manganese into
the steel. Finally, an imperfect admixture and an imperfect
influence of the ferro-manganese is, as a rule, impossible, as the
lively reaction which is produced proves.
The Co)iti)iitous Siccl Process in Fixed Furnaces 123
It is obvious that special quality steel is more easily made
ready in the furnace itself than in the ladle, but the above method
gives results leaving nothing to be wished for when producing
a hrst-class quality of soft ingot iron.
The ingot iron produced by the continuous process leaves
nothing to be desired. The same may be said as to the homo-
geneity of the material. As is well known, open-hearth ingot
iron is never absolutely homogeneous, so that one does not
speak of ideal homogeneity, but the metal made ready in the
ladle itself is quite as homogeneous as that made by the ordinary
process with the additions added in the furnace.
However, in order to be in a position to decide finally as to
the homogeneity of the product, a parallel series of carefully
conducted analytical experiments were carried out on ingots
made by the continuous and by the ordinary processes. The
experiments were made as follows : On the same day two charges
were tak:n, one from furnace which was made ready in the fur-
nace itself in the ordinary way, and the second from the continu-
ous furnace, which was deoxidized in the ladle by the addition
of ferro -manganese. These were analyzed and the results showed
that the differences in the composition are very nearly the same ;
in the continuous process the differences as regards the carbon
and manganese are rather less, and as regards the phosphorus
rather greater. In both cases the metal was very hot, and thus
the variations in the composition are not inconsiderable. This
latter circumstance, however, is in the present instance not so
important as the composition itself; let the variations be more
or less great, they are due, as is well known, to variations in tem-
perature and to other conditions. But these experiments have
a great importance, as they demonstrate that metal made (a)
by the ordinary process and (6) by the continuous process are
similar to each other and make it clear that the former fears
as to the metal deoxidized in the ladle being less homogeneous
than that produced by the ordinary process were perfectly
groundless.
The continuous process can naturally be still further de-
veloped, but in any case it has shown, during the last two years
of practical work, its capabilities and advantages. These ad-
vantages do not consist solely in the continuity of the process
itself, but in the longer life of the furnace, the higher production
124 The Iron and Steel Magazine
and yield, the lessened fuel consumption, and also in the sim-
plicity of the whole plant and of the whole practice. There may
be here and there some imperfections in connection with it, as
is the case with everything new, but these will, it is to be hoped,
be overcome with time.
COPPER ALLOYS *
SPECIAL BRASSES AND QUENCHING OF BRONZE
By L. GUILLET
Translated from the French for The Iron and Steel Magazine
{Concluded from page 2g, Vol. X)
Aluminum Brass — Theoretical Study. The characteristic
part played by aluminum in metallurgy is that of a deoxi-
dizing agent. Although no exhaustive study has yet been m^ade
of aluminum brass, it has been known for several years that
an addition of 2 or 3 per cent of aluminum to a brass of the
type Cu, 70 per cent, and Zinc, 30 per cent, gives a product having
nearly the strticture of forgeable bronzes and also a similar
strength.! In our experiments we started with two types of
brasses containing respectively 70 and 60 per cent of copper,
and to these were added from i to 10 per cent of aluminum, re-
placing in every case equal proportions of zinc. Aluminum was
added in the metallic form, most of it in the crucible before
melting and the balance immediately before casting. It was
found that with .5 to i per cent of aluminum and 60 per cent
copper the alloy asstimed a beautiful gold color and that this
color was retained up to 5 per cent of aluminum, when it became
pinkish. With 7 per cent of aluminum a superb rose color is
produced, appearing gray under a certain incidence of light.
With 10 per cent of aluminum the alloy is white. Up to 4 per
cent of aluminum the metal can be worked hot, but with a larger
proportion of that element working becomes very difficult, the
metal breaking if drawing be attempted.
The structure of these alloys is shown in Figs. 24 to 31. An
addition of .5 per cent aluminum to a brass containing 60 per
* " Revue de Mdtallurgie," February, 1905.
fCharpy, " Researches on Alloys of Copper and Zinc."
Copper Alloys
125
cent copper does not produce any structural change. With i
per cent aluminum the crystals of ZnCu are much more widely
separated. With 2 per cent of aluminum the crystals character-
FiG. 24. Cu,eo%;Zn,39.5%; Al, .5%.
Magnified 50 diam.
Fig. 25. Cu,6o%; Zn,36%; Al, 4%.
Magnified 50 diam.
Fig. 26. Cu, 60%; Zn, 34%, Al, 6%.
Magnified 50 diam.
t
<
iff .'■ <* I J.
Fig. 27. Cu,6o%; Zn,33%; Al,7%.
Magnified 50 diam.
istic of forgeable brasses are no longer found, being replaced by
polyhedric grains similar to those present in brass containing
less than 55 per cent copper. With 5 per cent aluminum poly-
126
The Iron and Steel Magazine
hedric grains are found containing small crystals similar to those
discovered b}-' Charpy in brass with about j 50 per cent copper.
With 7 per cent aluminum the grains are surrounded by a number
Fig. 28. Cu, 70%; Zn, 27%; Al, 3%.
Magnified 200 diam.
Fig. 29. Cu, 70%; Zn, 25%; Al, 5%.
Magnified 200 diam.
Fig. 30. Cu,7o%; Zn, 22%; Al, 8%.
Magnified 200 diam.
Fig. 31. Cu, 70%; Zn, 20%; Al, 10%.
Magnified 200 diam.
of dark areas. The alloy containing 10 per cent aluminum
was so brittle that it was not possible to polish it. Passing to
the brasses containing 70 per cent copper, when the proportion
Copper Alloys 127
of aluminum does not exceed 3 per cent the structure remains that
of ordinary brass. Upon exceeding that amount, large crystals
of forgeable brass appear in the structure. With 6 per cent of
aluminum polyhedric grains only are found, while with 10 per
cent of aluminum, the grains are found to contain numerous
small crystals.
To sum up, aluminum brasses have a structure similar to
those allo\'s containing a larger percentage of zinc. Their color
and behavior when worked lead to the same conclusions. For
instance, "svith 60 per cent copper, 2 per cent aluminum and 38
per cent of zinc, the structure is similar to that of a brass con-
taining 55 per cent copper and 45 per cent zinc. An addition
of 2 per cent of aluminum appears to have the same effect as
7 per cent of zinc, i per cent of aluminum being equivalent to
3.5 per cent of zinc. Further comparison justifies this conclusion.
It will be seen in the following pages that the same holds true
w4th regard to the mechanical properties conferred respectively
by I per cent of aluminum and by 3.5 per cent of zinc.
Manufacture. — Aluminum brass is obtained by melting
together copper, zinc, and aluminum. As the latter metal is
difficult to alloy, because of its low specific gravity, the molten
mxass should be mixed with care. A small amount of aluminum,
moreover, should be reserved for addition just before casting,
when it will act as a purifier. Instead of metallic aluminum a
copper alluminum alloy, prepared for that purpose, is frequently
used. The aluminum brasses generally used are of the following
two types:
. Cu 68 to 70%, Zn 31 to 27%, Al i to 3%.
Cu 59 to 61%, Zn 40.5 to 47-5%. Al .5 to 1.5%.
Properties. — Aluminvim brasses have a very close grain
and possess a tenacity greatly superior to that of ordinary brass.
The results obtained in our experiments are shown in diagrams
8, g and 10. From these results the following inferences may
be drawn: (i) Aluminum increases slowly the tenacity and the
elastic limit; (2) at first it increases the elongation, and then
decreases it; (3) when present in small amounts it increases
decidedly the shock resisting quality of the alloy ; (4) it increases
the hardness only when present in relatively large quantity, but
the alloy becomes then very hard.
128
The Iron and Steel Magazine
Uses. — Aluminum brasses are employed for the same
purposes as manganese brasses. It might be added that in
France aluminum brass has been used in the construction of
submarine boats, but it did not give entire satisfaction.
Other Special Brasses — Complex Types. To make these
experiments exhaustive the influence c»f some other elements
should have been studied, especially that of iron, vanadium,
nickel, cobalt, silicon, etc. These researches have been under-
taken and the results will be published in a few nionths. It
mav be stated now that the high tenacity brasses which are
wrongly sold as bronzes are very complex. They are made up
50
MS
uo
35
3o
55
Zo
/
^ .-
/
**--
"
\
•i
\
'
^0
0,5 1 i % ?
Diagram 8. Aluminum Brass. First type
of four, five and even of a greater number of constituents. They
may, however, be classified in two groups: (i) Those containing
from -5 to I per cent of manganese and a little tin, iron and alu-
minum, and (2) those containing 2 per cent of manganese and
a little tin, iron and aluminum.
Quenching of Bronzes. — In some very important experi-
ments Messrs. Heycock and Neville have shown, apparently
in a conclusive manner, what was the constitution of copper
tin alloys, that is, of ordinary bronzes. According to these in-
vestigations at the ordinary temperature, bronzes containing
over 75 per cent of copper may be divided into two classes:
(t) with o to 9 i,>er cent of tin they form a solid solution a, and
Copper Alloys
129
(2) with 9 to 3 1 per cent of tin a solution a and a definite com-
pound '5 answering to the formula Cu^Sn. The bronzes contain-
0,5 -1
Diagram 9. Aluminum Brass. Second type
0,7 1,i ^s z Q,,n
Diagram 10. Aluminum Brass
ing less than g per cent tin are always composed of the solution
a whatever their temperature, while with 9 to 22 per cent of tin
the bronzes change their constitution at 500° C, being then
I30
The Iron and Steel Magazine
composed of the solution o and of a solution /?, the latter contain-
ing from 22.5 to 27 per cent of tin. With a larger proportion
of tin, the change of constitution is more pronounced. It oc-
curred to us that these transformations might have an industrial
interest, and we were led to study the quenching of these alloys
by which the metal could be maintained in the conditions exist-
ing at higher temperatures. It should be borne in mind in this
connection that even now bronzes are occasionally quenched.
Mr. Rich, moreover, has shown that some bronzes containing
as much as 10 per cent tin could be rolled when heated to a cherry
red color while the}^ could not be rolled in the cold. In our
experiments samples of bronzes containing from o to 22 per
cent of tin were quenched at temperatures increasing between
300° and 800° C. The results obtained are tabulated below:
Type
I.
Cu =
= 79
Sn = 2 1
Quenching Temperature
Tensile Stren
Kilos per sq.
igth,
mm.
Elongation %
Reduction %
Not quenched
20.0
5-5
0
550
38.6
12.2
2.8
650
38-3
II. I
0.7
700
35-8
21.2
2.9
750
35-2
23.2
1-5
Type
II.
Cu =
= 84
Sn = 16
Quenching Temperature
Tensile Strength,
Kilos per sq. mm.
Elongation
%
Reduction %
Not quenched
25
4.7
1.4
300
22
?
0
400
24.4
?
0
500
19.4
5-6
1.4
550
40.1
10. 1
5-9
600
42.6
?
3-6
650
363
?
1.4
700
34-4
2.9
750
29.6
?
5
Type
III.
CU:
= 87
Sn = i;
3
Quenching Temperature
Tensile Strength,
Kilos per sq. mm.
Elongation
%
Reducticjn %
Not quenched
24.1
13-7
3
500
21.7
14.0
3
550
23-7
?
5
600
28.0
10.5
10
650
28.0
8.4
9
700
28.0
10.4
13
750
30.5
?
19
800
25-4
8.4
9
Copper Alloys
131
Typk
IV.
Cu ^ 9 1
Quenching Temperature
Tensile Strength,
Kilos per sq. mm.
Xot quenched
2 5-4
400
18.4
500
18.4
6co
25
700
25
8co
20.7
900
3-9
Type V.
Cu = 95
Quenching Temperature
Tensile Strength,
Kilos per sq. mm.
Not quenched
19.2
300
24.0
450
24.6
550
234
600
21.0
650
19-3
700
193
750
19.2
800
6.7
Sn
Elongation
%
R
?duclion %
10.3
16.5
10.5
14
10.5
1 1 -5
9.2
23-5
10.5
23-5
7-1
30
3-9
2
Sn = 5
Elongation
%
R
eduction %
?
20
10.3
27-5
I I.I
27
6.8
28
?
27
6.1
25
4.8
20
7-5
22
?
3
*^ ioo Zoo loo 1400 500 600 700 Soo aoo
Diagram ii. Type i. Cu, 79%; Sn, 21%
If these results are compared with those shown in'diagrams
II to 15, the following conclusions may be drawn:
I. In the case of alloys containing over 92 per cent copper,
the tenacity is slight!}^ increased by quenching between 400°
and 6co°, and the elongation is similarly affected.
132
The Iron and Steel Magazine
2. In the case of alloys containing less than 92 per cent
copper, the tenacity and the elongation increase decidedly as
HO
^o
Zo
iO
• V
V \
' \
/ \
0 100 200 500 "t'OO SOO ^00 'JOO $00 ^00
Diagram 12. Type 2. Cu, 84%; Sn, 16%
uo
30
lo
10
/
—7 \
/
/
0 100 loo loo uoo 5oo 600 "JCO %00 ^00
Diagram 13. Type 3. Cu, 87%; Sn, 13%
soon as the quenching temperature exceeds 500°. This is in
full accord with Messrs. Heycock and Neville's diagram.
Maximum strength is reached, whatever the composition, at
a quenching temperature of about 600 degrees.
Copper Alloys
133
Maximum elongation on the contrary seems to be affected
by the composition of the alloy. With 91 per^cent copper,
maximum elongation corresponds to a quenching temperature
Uo
3o
Zo
to
3~-T
^*"*^ J ' V '
-^fc^ — 1 -y- 1 V 1 —
' \ '1
' \ 1
' ^ ~ - ' \ 1
0 -100 loo ^0Q Mco Soo Soo ?oo 5oo 500
Diagram 14. Type 4. Cn, 91%; Sn, 9%
ho
^ .*•<
'—- .
,j
*■*
■,^
^-.
,.•
>
»
••"
^,
""^^^
2p
.^^^■^
■
""""''^
>
V
(
N
AO
>
\ «
■\ > —
0 100 2oo loo Hoo Soo 600 700 800
Diagram 15. Type 5. Cu, 95%: Sn, 5%
of 800° while with 79 per cent the maximum elongation corre-
sponds to a quenching temperature of 600°. The difference
between the tenacity of the cast alloy and that of the metal
134 The Iron and Steel Magazine
quenched at the most desirable temperature, is the greater
the smaller the percentage of copper. It may be concluded
from these results that the quenching of bronzes causes a decided
improvement in the results obtained by tensile tests. It would
be interesting to ascertain the effect of quenching upon the fric-
tional properties of bronze, for it m.ust be rememl^ered that
quenching causes the disappearance of the definite compound,
a very hard substance, which must play an important part when
the bronze is employed for bearing purposes. The writer hopes
soon to investigate this point.
VANADIUM AND VANADIUM STEEL*
np HE increasing importance of vanadium to the modern steel
maker will render a resume of the methods by which this
metal has been obtained from its ores of interest to readers of
" The Engineer." Of the rare metals which, alloyed with iron
in steel, have most important effects upon the physical properties
of the latter, vanadium seems to be the one by which the greatest
effect is produced with a minimum addition of the element.
There are, in fact, some well-known metallurgists who assert
that the properties of Swedish iron and steel are due to the pres-
ence of vanadium oxide in the original iron ore from which this
iron and steel is made; and the special characteristics of the
latter are ascribed to the very minute percentage of vanadium
which finds its way into the finished material, and is generally
overlooked in the analysis of the same. There are, of course,
other metallurgists who hold this view to be incorrect, and who
believe that Swedish iron and steel owes its superiority to other
causes.
However, whether this theory regarding Swedish iron and
steel be correct or not, vanadium steel undoubtedly possesses
physical characteristics which render it suited for special require-
ments. The following description of the methods by which
vanadium and its alloys can be prepared, from vanadium ores,
and the resume of the researches by Guillet and others upon the
physical properties of the vanadium steels made from these^
alloys should therefore prove of value to modern steel makers.
* " The Engineer," June 9, 1905.
Wuhidiuni and Vanadium Steel 135
The presence of vanadium in a lead ore from Mexico was
discovered by Sefstrom, a German chemist, in the year 1830,
and the properties of the supposed new metal were investigated
bv Berzelius. the noted Swedish chemist, in the following year.
Roscoe. in 1867, however, showed that the compound Berzelius
had supposed to be the metal was in reality the oxide of the
nitride, and that the properties he had ascribed to it were thus
incorrect — the metal being allied in properties and character-
istics to the arsenic and phosphorus group of elements, and not
to the chromium group as Berzelius had supposed.
Vanadium occurs in nature chiefly in the form of com-
pounds of the higher oxide V2O5, the acid being known as vana-
dic acid, and forming salts represented by the general formula,
H3VO4, and called " vanadates." On loss of one molecule of
water the vanadic acid becomes metavanadic acid, and the
general formula for salts of this acid is H VO3. Vanadium occurs
in nature in many of the ores of lead and iron, and at one time a
vanadium-bearing lead copper ore was mined in Cheshire. As
already pointed out in the introduction to this article, the special
properties of Swedish iron and steel are now attributed by some
metallurgists and engineers to the presence of vanadium in the
original ore from which it is smelted.
Pure vanadium was prepared by Roscoe in the chemical
laboratory by heating the dichloride V2CI4 in a current of hydro-
gen gas. The hydrogen combines with the chlorine and carries
it off as hydrochloric acid gas, while the metal remains as a light
gray metallic powder, having a specific gravity of 5.5. This
powder does not oxidize or tarnish when exposed to the air, but
it melts at about 1700° C, and when strongly heated in air it
combines with oxygen and yields V2O5.
Minerals containing vanadium are found in Spain, Mexico,
Sweden and Scotland, and probably now that the metal has
become of considerable commercial importance, its presence
in a large number of minerals and in various localities will be
demonstrated.
Extraction of Vanadium from its Ores
Several methods of extracting vanadium from its ores have
been proposed. Some of them are now in use upon an industrial
136 The Iron and Steel Magazine
scale of operations, and the following notes give briefly the
details of the more important of these :
( 1 ) At the works of Bas Coudray , near Le Genest, in France,
a Spanish ore containing lead vanadate is employed. This
ore contains 12 to 14 per cent V2O5 and about 50 per cent Pb.
The ore is first smelted for its silver and lead contents, with
carbonate of soda and coal. The lead and silver are thus sepa-
rated in metallic form, and the vanadium passes into the slag
as sodium vanadate. The slag is then submitted to a melting
temperature in another furnace, and air is blown over it to oxi-
dize the vanadium and form vanadic oxide, V2O5. The molten
slag is then granulated by running it out of the furnace into boil-
ing water, and from this the vanadic acid is extracted by leaching
in suitable tanks. Silica is removed from this solution by careful
addition of sulphuric acid, and the vanadic oxide is finally pre-
cipitated by use of an excess of the same reagent. By use of
ferric sulphate, or sulphates of other metals, in place of sulphuric
acid, a mixture of vanadic oxide and ferric or other oxide can be
precipitated. From these mixed oxides special alloys of vana-
dium and iron, nickel, copper or cobalt can be prepared by the
Goldschmidt aluminum reduction process. Further details of
this method of obtaining vanadium will be found in the original
memoir by Herrenschmidt in " Comptes Rendus," 1904, p. 635,
Vol. CXXXIX.
(2) Gin, the French electrometallurgist, has proposed a
method of extraction which, in some respects, resembles that
used for the electrolytic reduction of alumina ; but in this case
not pure vanadium, but an alloy of vanadium and iron, is pro-
duced. A bath of molten calcium and ferric chloride is prepared,
and this is electrolyzed with an iron cathode and an anode com-
posed of an intimate mixture of vanadium trioxide and carbon.
The fluorine which is liberated at the anode attacks this oxide
and forms vanadium fluoride, and this in turn is decomposed
by the current with liberation of vanadium at the cathode. This
cathodic vanadium then combines with the iron produced by
the decomposition of the ferric fluoride, and an alloy of iron and
vanadium is finally obtained in the molten state on the floor of
the bath. A current density of six amperes per square centi-
meter at the cathode is required with an electromotive force of
10 to 15 volts.
VinnuUu))! ami Vauadiuni Steel 137
The following chemical equations represent the three steps
in this process of manufacturing ferro vanadium alloys by elec-
trolysis :
(i) FejFe = 2 Fe + 6F.
— +
at the at the
cathode anode
(2) V,03 + 3 C + 6 F = 2 VF, + 3 CO.
(3) 2 VF3 = 2 V + 6 F.
— +
at the at the
cathode anode
Further details of this method will be found in the paper
read by M. Gin before the Fifth International Congress of
Applied Chemistry at Berlin in 1903.
(3) Carpenter, an American metallurgist, has patented the
following method of smelting ores containing vanadium trioxide,
in order to obtain ferro vanadium alloys. The ores, which may
contain .5 to 5 per cent of vanadium, and much silica, alumina,
etc., are smelted by the ordinary blast-furnace procedure, in the
presence of a sufficient amount of iron oxide, to produce the iron
required for the final alloy. The fuel and blast must be arranged
to produce a very high temperature together with a very strong
reducing action, and the iron oxide must be intimately mixed
with the vanadium-carrying ore, in order to insure that iron is
present at each point where vanadium is liberated. Dolomite
or some other basic flux is added to slag off the silica, when this
is present in ver^^ large amounts. This method is covered by a
United States patent, No. 781,808 of 1905.
(4) Goldschmidt uses the following method for preparing
alloys of vanadium and aluminum. The oxide of vanadium,
perfectly dr)% is intimately mixed with powdered aluminum in
calculated amount. A small portion of this mixture is now
introduced into the type of reducing crucible used for this work,
and the reaction is started by a sodium-peroxide cartridge and
a red-hot iron, in the customary manner. The remainder of the
charge is then projected in small portions into the crucible
as the reaction proceeds and an alloy of vanadium and alumi-
num is produced under the slag of fused alumina. The propor-
138 The Iron and Steel Magazine
tion of vanadium in this alloy can be varied at will, but one
containing 20 per cent of aluminum is the most convenient for
use.
(5) Moissan attempted to reduce vanadium oxide directly
by carbon, in his early experiments with the electric furnace.
The reduction took place very slowly, and even with a current of
1,000 amperes at 70 volts an impure product containing much
carbon was obtained.
Details of this experiment will be found in '' Comptes Ren-
dus," Vol. CXVI, p. 1225, and also in Moissan's classical work
on the electric furnace.
(6) Von Bolton, in the laboratory experiments for Messrs.
Siemens and Halske which have resulted in the new tantalum
incandescent lamp, carried out some investigations upon the
reduction of vanadium trioxide which may be referred to here.
The brown oxide powder was intimately mixed with paraffin
and was made into rods, which were then embedded in granular
carbon and heated to a temperature of 1700° C. The hardened
and compact rods of vanadium trioxide, .8 millimeters in diame-
ter, which remained after this treatment, were then placed in a
vacuum apparatus and were electrically heated with a current
of 1.8 amperes at 42 volts. Much oxygen was given off as the
rods became white hot, and this was carried off by continuing
to work the vacuum apparatus. The metallic gray mass which
remained was composed largely of pure vanadium, and had a
melting point of 1680° C. This metal was not used, however,
for the new lamp, since its melting point was 550° C. lower than
that of tantalum.
Properties of the Vanadium Steels
I. Guillet, the well-known French chemist and metallurgist,
has published several researches bearing on the properties of the
vanadium steels. The results of the first of these were presented
to the French Academic des Sciences early in 1904, and were
published in " Comptes Rendus " of February 8, 1904. They
dealt with two series of steels containing a variable percentage of
vanadium, with high and low carbon contents. The addition
of vanadium was found to alter the physical structure of the
steel, the results being as follows:
]'\i]hidii())i ami Vauadinin Steel
139
No. I Series
No. 2 Series
Structure
Carbon .10 to .20 per
cent
Carbon .60 to .85 per
cent
Pearlile onlv
Up to .70%
vanadium
From .70 to 3.0%
vanadium
From 3.0 to 10.0%
vanadium
Up to .50%
vanadium
From .50 to 7.0%
vanadium
From 7.0 to 10.0%
vanadium
Pearlite and double carbides
Double carbides only
The pearlite steels were not more fragile than ordinary steels,
but were harder, while the double carbide steels were low in
tensile strength and brittle. The steels containing both pearlite
and double carbide occupied an intermediate position, as regarded
strength and hardness.
A steel containing 1.04 per cent vanadium and .112 per
cent carbon gave the highest tensile strength in the first series,
namely, 61.1 kilos per square millimeter, while one contain-
ing 4.99 per cent vanadium and 1.084 per cent carbon gave the
highest strength test — 98.9 kilos per square millimeter — in
the second series. In a further commtmication to " Comptes
Rendus," dated August, 1904, Guillet states that the steels
containing a high percentage of vanadium are of particularly
irregular strength and structure, as the light carbide of vanadium
tends to float when the metal is being cast.
The general conclusions of Guillet, based upon these series
of tests, were that only those steels containing less than .70 per
cent of vanadium could be considered of industrial value. The
brittleness of these steels, however, rendered them unsuitable
for use in the manufacture of cutting tools, and, in the author's
opinion, another series of investigations were required to decide
whether this brittleness could be overcome by the use of nickel
with the vanadium. This investigation Guillet intended to
undertake when opportunity offered, but up to the present no
results of this new investigation have been published.
2. Capt. H. R. Sankey and Mr. G. K. Smith, in December,
1904, read a paper before the Institution of Mechanical Engineers,
upon " Heat Treatment Experiments with Chrome-Vanadium
I40 The Iron and Steel Magazine
Steel," which contained much valuable information relating to
the effects of heat upon steel containing vanadium and chro-
mium. This paper and the discussion upon it were fully reported
in " The Engineer," of December 23, 1904. The authors used
in their experiments a raw chrome- vanadium steel, the exact
composition of which was not stated. This steel showed a ten-
sile strength of 54 tons per square inch, and, at the same time,
behaved under bending and twisting stresses like mild steel of
the highest quality. Its power of resisting torsional strain
was also stated to be enormous — a piece 6 inches in length and
.75 inches in diameter twisting 3.9 times before parting.
3. A series of tests haA^e been made b}^ Professor Arnold of
Sheffield, and other metallurgists, upon the steels produced
from the special ferrovanadium alloys made by the New Vana-
dium Alloys Compan}', Limited, of London, and of Llanelly in
South Wales. The results of these tests have been published in
" The Engineer " of July i and September 2, 1904, in letters
signed by A. F. Wiener, the managing director of the company.
The most important tests of this series showed that the
tensile strength of plain steel containing .25 per cent C, and
.40 per cent manganese was raised from 30 tons to 47 tons per
square inch by the addition of .25 per cent vanadium, and that
a similar addition in the case of a nickel steel raised the tensile
strength from 42 to 68 tons per square inch.
The highest figure attained in these tests was for a nickel
vanadium steel of the following composition: .141 per cent C,
.512 per cent manganese, 9.36 per cent Ni., and .29 per cent V.
This stood 101.20 tons per square inch before parting in the
testing machine, as compared with 88.8 tons, the previous record
for nickel steel.
As a result of the above test. Professor Arnold stated that
" it was demonstrated be^/ond doubt that the addition of a few
tenths per cent of vanadium raises the elastic limit of mild struc-
tural steel at least 50 per cent without seriously impairing its
ductilitv."
s
Metal Mixers jor Pipe Foundries 141
METAL MIXERS FOR PIPE FOUNDRIES *
By J. B. NAU, New York
OME time ago the writer had an opportunity to take a trip
through some of the most modern European foundries and
noticed that in many respects these foreign foundries are ahead
of American practice. This was more especially to be seen in
pipe foundries. The pipe specifications of Europe, while very
strict and exacting in many ways, in fact more so than in the
United States, leave the manufacturer some latitude not en-
joyed here. This is especially true in regard to the iron used..
Whenever possible, for obvious reasons, the European manufac-
turer makes use of direct metal from the blast furnace. Of
course every precaution is taken always to obtain an even
quality of iron by mixing the metals running from the different
furnaces in special ladles, or in mixers of special design that can
be nm from one furnace to the other. The principal aim of the
manufacturer is naturally to give full satisfaction to the buyer,
who, as is proper, insists that the pipe bought shall fill all the
conditions of the specifications in regard to ph3^sical tests and
strength of the metal. The buyer on his side insists simply on
the pipe coming up to the specifications and leaves the manu-
facturer unhampered as to the source from which he should take
his iron.
From long experience the manufacturer is well aware that
he can satisfy all physical tests with iron whose analysis may
vary within certain limits and that he can obtain the right qual-
ity either by mixing the iron from different blast furnaces or,
if necessary, by correcting this direct metal by admiixing a cer-
tain amount of metal melted in the cupola. This item of quality
is generally very carefully attended to and the pipes made,
especially in some of the most improtant French and German
plants, where the best working methods in other respects are
employed, is invariaVjly of good quality.
It can be readily understood that the use of liquid metal
directly from the blast furnace is fraught with many difficulties
and its indiscriminate use is liable to produce pipes of uneven
* A paper read before the American Foundrymen's Association, New
York, June, 1905.
142 The Iron and Steel Magazine
and uncertain quality. That this is so is well illustrated by the
fact that the American pipe specifications, in order to eliminate
the factors of uncertainty and of uneven quality, prohibit the
use ot direct metal in the manufacture of pipes, and specifi-
cally state that the iron should be remelted in a cupola or an air
furnace. Of course this remelting leaves no excuse to the manu-
facturer in regard to the quality of his iron, but still it has disad-
vantages that cannot be overlooked, one of which is the greatly
increased cost of the iron when thus remelted. Another dis-
advantage lies in the danger of deteriorating the quality of the
iron to some extent b}' increasing its sulphur content from the
contact with fuel or flame. The best and cheapest way would
naturally be to follow the practice already described, by mixing
the different brands of irons running from the various blast fur-
naces and, if necessary, to correct the quality of the irons so
mixed with irons remelted in the cupola. Such a proceeding,
prohibited by American specifications, is only possible where
the foundry is built near the blast furnace. Foundries not so
located are naturally excluded.
vStill, in order to get all the benefit that the use of direct
metal would undeniably afford anci yet avoid all the danger of the
uneven quality, that alone guided the framers of the American
•specifications in prohibiting the use of direct metal, the use of a
heated mixer of sufficient capacity is here recommended. A
few years ago such mixers were introduced in Europe in connec-
i:ion with steel works and they give the very best satisfaction.
The writer had occasion to see one such heater of a capacity
of 250 tons in a basic Bessemer steel works. It was heated with
producer gas and provided with regenerators. Such a mixer is
more to be considered in the nature of a gas-heated air furnace,
■with the exception that the iron, instead of being melted in it,
IS brought liquid from the blast furnace and poured into the
mixer, where it can be kept hot and liquid for any desired length
of time.
Heated Mixers Preferred
Since the steel works find that heated mixers give such good
satisfaction, why should the foundrymen not follow in the
same steps wherever it is possible to do so? The small non-
heated mixer that the writer saw in some German pipe foundry
Metal Mixers for Pipe Foundries 143
gave good satisfaction and the iron could still be used after two
hours' waiting. Heated mixers would give better satisfaction
undoubtedly. Under the action of the flame, which must be of
a slighth' ON:idizing nature, some undesirable refining of the
iron would take place and a small part of the silicon and man-
ganese would be burned out. But as the iron is poured liquid
into the mixer the oxidation of silicon and manganese would
be much less than in the air furnace or the cupola, where from
one quarter to one third of the silicon is burned out, mostly
during the melting period. In this respect the heated mixer,
where no melting is done, should be far more advantageous.
But it has other advantages. It is a well-known fact that
the mixer as such is an excellent apparatus for eliminating in
a short time a large percentage of the sulphur contained in the
iron. In fact, this elimination reaches easily 50 and 70 per cent.
The quality of the iron will, therefore, be considerabl}^ improved
in the mixer. In the cupola the contrary always happens.
Since the metal can be kept hot to any degree in the mixer,
this affords us means to correct at once the quality of the metal
by throwing and melting in it whate\er cold pig may be neces-
sary to obtain the right analysis.
Thus the use of the heated mixer for foundry purposes in
connection with the blast furnace would seem to present very
marked advantages :
1. It would allow the use of direct metal from the blast
furnace, thereby saving the remelting cOvSt, which is always high.
2. It would afford an easy and cheap means of elim.inating
a large percentage of sulphur, thereby, contrary to cupola prac-
tice, improving the quality of the iron.
3. It could be used directly to correct the quality of the
metal by melting in it a certain amount of cold pig iron, with
probably less cost than when the melting is done in the cupola
and with certainly less absorption of sulphur during the melting,
because the solid pig w411 melt in the liquid bath and be removed
from contact with the flame.
Naturally such a mixer could be used for other than pipe
foundr\^ purposes when proximity to the blast furnace allows
its adoption. It would even allow an independent large foundry
to use liquid iron from a nearby blast furnace belonging to an-
other company. Both companies would benefit bv the practice.
144 ^^^ Iron and Steel Magazine
Features of European Pipe Foundries
Coming back again to Enropean pipe foundry practice,
which led us to the above consideration of mixers, many other
improved methods are applied, all tending toward a general
cheapening everywhere. What one notices first in the most
modern European pipe foundries is the complete absence of pits.
The molding and pouring are done on an upper floor to which
one half of each pipe flask is fastened ; the other half is left free
to be Slid back. The flasks are never taken from their place.
The finished pipe alone is lifted up by the crane, which having
less weight to lift can be lighter than would otherwise be the
case. The lower end of the flask hangs freely down to within a
certain distance of a lower floor and is accessible from every
side. The drying of the flasks is done by means of producer
gas or even blast-furnace gas. The producer gas is made in gas
producers situated outside the works and let through under-
grotmd gas flues underneath the rows of flasks. Here special
burners, varying in the different foundries, are used to dry each
flask thoroughly and rapidly. In most places gas is also used
to heat the core ovens. These arrangemients also vary in difter-
ent places.
In one foundry in eastern France, where coke is very expen-
sive (fullv $5 a ton), special arrangements are made to burn
coke screenings from blast-furnace coke in the core ovens. The
use of this fine coke, otherwise lost, was the cause of a very
great saving in the drying of the cores. After the pipe is cast
the sand falls from the fl^ask and is taken up by conveyors, and
after being cleaned and screened is taken by conveyors to the
top floor, where it will be used again in the ramming of the flasks.
In some foundries the flasks are put up exclusively in straight
parallel rows ; in others, especially for pipes of less than 25 inches
diameter, revolving pits or drums are used, somewhat similar
to what is done at the Chattanooga plant.
In the case of revolving systems the drying of the flask is
also obtained with gas. Some very ingenious drying appliances
are used in these systems. Tarring appliances also are very
modern and of an improved style. In fact, the work everywhere
is reduced to a minimum.
T
Alcliiiif^ Steel with Cast Iron 145
MELTING STEEL WITH CAST IRON*
By R. P. CUNNINGHAM, Holyoke, Mass.
PIE demand for castings to stand great strains has increased
to such an extent that fotmdrymen are often at a loss
how to produce castings up to the required specifications. The
manufacturers who are the most often called upon to produce
castings of high strength are pump and engine builders, tool
makers and car-wheel manufacturers. With pump builders
a few years ago it was something very unusual to receive an order
for a pump to stand a pressure of more than 1,000 pounds. To-
day it is nothing uncommon to get an order for a pump to work
under a pressure of 5,000 poimds, and even higher. Engine
builders are called upon to build engines to work under 200
pounds steam pressvire, while only a very few years ago too
pounds pressure was considered abotit the limit.
I miight say the same thing about tool making. The speed
at which the modem tools are run to-day is nearly double that
of a few years ago. Look at car wheels and compare the tests
they are subjected to to-day with those required twenty-five
years ago. The increase is over too per cent. Yet car wheel
makers have managed to make wheels that come up to the re-
quirements. I might go on and enumerate many other branches
of the trade that are doing what was once considered an impos-
sibility. This goes to show that the foundrymen of to-day are
alive to the requirements, and yet we often hear men say that
the foundry has not progressed as fast as other branches of manu-
facturing. On the contrary, considering the attention that has
been paid to the foundry, we have managed to make castings
that have been far above the specifications called for. Foundry -
men do not always have the iron in their yards to make castings
of any required strength, but by a judicious use of steel scrap
we can produce castings of the strength desired.
Any one familiar with pump work will readily understand
the necessity of having a perfect casting, not alone smooth and
true to pattern, but clean, close grained, yet soft enough to ma-
chine easily. Many castings go through the machine shop and
* A paper read before the American Foundrymen's Association, New
York, June, 1905.
146 The Iron and Steel Magazine
erecting room, but fail when put under test. This adds to the
manufacturing cost, as often the machining is many times the
cost of molding. By adding a percentage of steel scrap we
have in a great m.easure overcome this difficulty if the trouble
is caused by porosity of the metal.
Care in Charging the Cupola
When melting steel with cast iron there are many things
that require close attention in order to obtain the very best
results. In charging the cupola one cannot be too careful and
should be absolutely certain that all the material called for in
the charge is put in. The weight of each material specified
should be correct; the fuel and fluxes should be analyzed so
that the exact composition of all the materials going into the
iron to be made may be known.
In making high-grade metal we have to contend with the
impurities of the fuel and fluxes charged into the cupola besides
that we have estimated on in the metal. All impurities in excess
tend to weaken the metal in tensile and transverse strength ;
for this reason there is more difficulty in making a successful
cast when using a large percentage of steel scrap.
A high percentage of steel necessarily increases shrinkage,
demands closer attention, requires more rapid handling in the
foundry and when very high tends to make all the operations
connected with it draw away fromi those of a cupola metal
and approach that of a steel casting. When this extreme point
is reached melting in the cupola becomes very unsatisfactory.
The average thickness of a casting bears a relation to the
percentage of steel desirable. For thin castings only a small
percentage can be used, while for thick, heavy castings a large
per cent is permissible. This is so because a thin casting has no
self-annealing power, on account of its rapid cooling and the
chilling effects of the m^old. The thicker casting, on account
of its slower cooling, anneals itself somewhat and opens the
grains of the metal perceptibly. The same metal in a thin
casting, which is hard, would be quite soft in a heavy casting.
My opinion is that it is more desirable to have a mixture with
the smallest percentage of steel that will give sufficient strength
and solidity to the casting for all practical purposes.
Mcliiiii!, SiccI with Cast Iron 147
We sometimes doubt the wisdom of the engineer when he
calls for castings that will stand so many thousand pounds to
the square inch, because the metal that will stand the highest
test in the bar is not always the most desirable. It may be
brittle or flaky, with no elastiii^ity, and yet test high. What
we aim for in practical foundry work is a high-grade metal that
will stand a fairly high test and machine easily. It is this kind
of a casting that can be made with a percentage of steel scrap
melted with your iron, provided the rules are accurately followed.
My method of charging a cupola is as follows: Let us say
that we want to make a casting which will require 4,000 pounds
of metal, with 25 per cent steel. With a cupola that lines up
48 inches, we put on the bed 1,200 pounds of coke, on top of this
put 1,000 pounds of iron, then 500 pounds of steel, then 500
pounds of iron, then 150 pounds of coke, 500 pounds of steel,
1.500 pounds of iron. The coke next above the metal charge
should be greater than between the ordinary charges, and the
pig iron in the next charge above the steel should be of the same
chemical analysis as the iron used in the steel, so that if any metal
should melt and run into the steel it will do no harm. With
the last amount of steel we add ij pounds of ferro-manganese
to every 100 pounds of steel used. We also put the same amount
of ferro-silicon into the ladle. This should be done after the
first metal has been drawn into the ladle. This metal should be
poured as soon as it becomes quiet in the ladle.
If the casting is uneven in thickness attention must be given
the shrinkage. Setting a riser on the heavy parts and after
the mold is full pouring slowly until the riser is full obviates
trouble. If the casting is very heavy it will be necessary to
feed it, but an ordinary casting will not require this.
Selkctiox of Pig Irox axd Scrap
We have found by using two brands of iron, one high in
manganese and the other high in silicon, both low in sulphur,
that we can get a much finer grained casting, with more elasti-
city, than we could if we depended on ferro-manganese and ferro-
silicon to bring these two elements up to the desired percentage.
I reason it in this way: If the manganese and silicon are in the
pig they are more evenly distributed than when they are put
148 The Iron and Steel Magazine
into the cupola, and depended upon to become thoroughly mixed
in it or in the ladle. We have never yet depended upon the pig
for the entire amount of manganese or silicon wanted, but have
added each in the proportion given above. We sometimes have
trouble caused by wrought scrap or hard steel becoming mixed
v/ith the steel scrap. In either case satisfactory. results cannot
be obtained. With hard steel there are hard spots in the casting,
while wrought iron increases porovisness, which is very bad if
the casting is uneven in thickness.
My opinion is that mixtures of this kind will be used in the
future to a greater extent than in the past because the demand
for this class of castings has increased and foundrymen will
readily see that by this means they can build up their present
mixtures to show greater strength and other desired qualities.
The result of 18 casts with different percentages of steel
showed that the highest amount of steel that could be used to
advantage is 33 per cent. Above this showed excessive shrink-
age and only slight gain in strength. The highest point reached
for tensile strength was 33,205, the lowest 31,890, for perfect
bars. The highest transverse strength shown was 3,335, the
lowest for a perfect bar was 3,180. Six bars were cast from each
heat, two at the first part, two in the middle of the heat, and two
at the end. In every case the two bars cast in the middle of
the hea.t showed up best in tensile and transverse strength. The
first bars were not uniform and showed small pin holes. The
last bars showed up badly in every instance. Less trouble will
be had with less than 33 per cent than above that amount of
steel scrap in the gray iron mixture.
For ordinary work 25 per cent of steel will give sufficient
strength for all practical purposes, will machine easily and yet
be close grained. This is the per cent I would recommend
foundrymen to use unless it is for some special work.
Rail Sections as Ejif^iiiccritig Structures 149
RAIL SECTIONS AS ENGINEERING STRUCTURES *
By P. H. DUDLEY
'T'TTE mechanical properties, as stiffness and strength of a
section, increase in a rapid ratio as the height is augmented,
as shown bv Table No. i.
Ta
BLE No.
I
Weight of
Section
in pounds
Height
in
inches
Width
of
Head
Moment of Inertia,
fourth power,
inches
Moment of Re-
sistance, cubic
inches
60
4.0
2i
12.0
6.7
65
4-5
2f
16.0
7.8
80
5^
2H
28.3
II. 4
TOO
6.0
3.0
48.5
16.6
The 80- and 100-pound sections become more efficient
engineering structures than the 60- and 65-pound which they
replaced, by inducing a longer distribution of the passing wheel
loads to the cross-ties and ballast, and lessening the deflection
under the wheels. A greater portion of the wheel effects is ab-
sorbed by the constraining negative bending moments in the
wheel spacing, reducing the positive moments under the wheels.
This favorable action for a smoother running surface in the
general depression under the wheels, however, increases the
wheel contact pressure intensit}^ per square inch in the bearing
surface of the section, and imposes a greater burden upon the
rnetal of the entire head.
The stiff sections for a given unit fiber strain carry larger
bending moments than the limber rails, and therefore are more
efficient engineering structures for heavier axle and total loads.
Table No. 2
positive bending moments in inch-pounds for the given unit
fiber strains per section in table no. i
Section 10,000 lbs. 20,000 lbs. 30,000 lbs. 40,000 lbs.
60 lbs. 67,000 I. lbs. 134,000 I. lbs. 201,000 I. lbs. 268,000 I. lbs.
65 ,, 7^,000 ., 156,000 ,, 234,000 ,, 312,000
80 ,, 114,000 ,, 228,000 ,, 342,000 ,, 456,000 ,,
100 ,, 166,000 ,, 332,000 ,, 408,000 ,, 664,000 ,,
* Presented at the June, IQ05, meeting of the American Society for
Testing Materials.
150 The Iron and Steel Magazine
In the 80- and 100-pound sections nearly the given maximum
moments have been obtained in tests.
Unit fiber strains in tension from .0 to 30,000 pounds are
those which occur daily under present locomotives, with some
strains of the higher figures and liable from a flat wheel to be
exceeded.
Large unit fiber strains have been common since steel has
been used for rails, for the limber sections had frequent sets,
which indicated they were strained beyond their elastic limits.
Under 65-pound rails in a yard track, a unit fiber strain
has been meastired of 56,000 pounds.
The elongation for the unit fiber strains in Table No. 2
would be .00034, .00067, -ooi ^^^^ -00134 of an inch respectively,
for the modulus of steel at 30,000,000 poimds, summer tempera-
tures, though the modulus is higher in the winter.
The height of the section is increased to augment the mic-
chanical properties of stiffness and strength. This has a tend-
encv to reduce the unit fiber strains in the base of the rails,
but the .subsidence of the roadbed under the wheel loads, the
looseness of the cross-ties in the ballast, and the rails under the
spikes, cause the bending nioments carried by the stiff rails to
exceed those by the limxber sections for the same wheel loads,
though the percentage distributed per individual cross-tie is
less. This reduces the area of contact and increases the inten-
sity of the wheel pressures per square inch in the bearing surface.
The wheel loads also have been doubled on the stiff rails, over
what they were on the light sections, and the metal in the bear-
ing surface, therefore, sustains two and three times the burden
required by the former limber sections. (See Vols. Ill and IV
of the Proceedings, for Unit Fiber Strains and BendingMoments.)
To carry the trains and distribute the wheel loads, the rail
section has the top of the head or bearing surface shaped for the
wheel treads, to receive their pressures and distribute the loads
to the cross-ties, ballast and roadbed. The side of the head of
the rail is the gtiide for the passing wheel flanges, while the
entire section becomes the girder to distribute the wheel loads.
The metal in the rail section has three functions to perform,
to receive support, guide and distribute the wheel loads of the
passing trains. First, the metal in the bearing surface sustains
its loads principally by its properties of cubic elasticity, and
Rail Scciions as Eiii^^iiiccriiii::, Structures
151
should be sound and homoi^cneous. Second, the side of the head
resists abrasion of the wheel flanges by its toughness and tenacity.
Third, to distribute the loads by the entire section its linear
elasticity is exercised.
The factor of safety in the girder determines what physical
properties may be used for the bearing surface and guide.
The metal in tlie head of the rail must receive and sustain
the wheel contact pressures by its cubic elasticity, but distri-
butes the loads through the section as a girder, by its linear
elasticity.
Fig. I. A Rail Section from the upper portion of the ingot with a
central core of nieta.1 in which by liquation the metalloids in the steel are
above the average. A central pipe is also indicated, which may be formed
without decided liquation.
The distortion of the rail head in service shows that the
steel has low limits of cubic elasticity, and is not homogeneous.
(See Figs, i and 2.)
When the steel is solid, sound and of fine texture, the head
does not become distorted under the service, though it wears
in the bearing surface and on the side. When the ingot is un-
sound and spongy, then the head flattens and crushes under the
wheel treads. Rails from the top of the ingots, where by liqua-
tion the upp'er portion contains a higher percentage of carbon
and phosphorus, the central core of metal is not soimd and
strong, but fragile, and does not sustain the wheel contact pres-
sures as well as the exterior portion of the section. (See Fig. i.)
152
The Iron and Steel Magazine
Fig. 2. Check developed in head
of rail, twelve feet in length, after
service. Sound at the ends.
When a decided pipe in the ingot did not occnr in cooling,
the repeated pressures of the wheel contacts develop a check
which is equivalent to a pipe, the metal immediately over it in
the bearing surface stretching
sidewise by its linear elasti-
city ; the check widens until a
portion of the head becomes
detached from the web of
the rail, unless rem.oved from
the track. (See Fig. 2.) The
steel in the head is not homo-
geneous, either in quality or
structure, and becomes dis-
torted as a section, from in-
adequate physical properties
of cubic or elasticity of vol-
ume, to sustain and distribute
the wheel loads.
The splitting of the head
in the earlier steel rails was in nearly all cases traced directly
to a pipe in the ingot. These conditions still exist, yet there
are numerous instances in which the pipe did not develop in
cooling, but does in service, in the unsound metal of the central
core of steel, as indicated in Figs, i and 2. Pieces break from
the side of the rail head, in steel where so decided liquation has
occurred in the ingot as to make two or more grades of steel in
the rail head.
The changed relations of the steel are not appreciated in the
bearing surface to the wheel contact pressures in the stiff sections,
as the efficient engineering structures which have empowered the
present wheel and total loads of the locomotives and cars. By
the design of the section for a given weight, its mechanical
properties have been increased to transport heavier loads. To
complete it as an engineering structure, with the requisite effi-
ciency, the physical properties must also be augmented in sound
metal, to raise the limits of its cubic elasticity in the bearing
surface proportionately to those required by its enlarged mechan-
ical properties as an engineering structure, — a question of me-
chanics, involving a metallurgical solution.
Iron rails failed in the bearing surface when the wheel loads
7^c7/7 Scciiois as E}iiii)iccri)ii; Structures 153
increased to 10,000 pounds, owing to low limits of cubic or elas-
ticity of volume of the metal. The two or three per cent of slag
made it a bundle of fibers only for the wheel pressures, though
adequate to distribute the loads as a girder. Increase in stiffness
in the iron rails to carry larger bending moments hastened their
destruction.
In the limber steel rails first rolled, the grain or texture of
the metal was fine and compact; the elastic limit in reference
to its cubic elasticity was high, and could sustain the wheel
contact pressures without distortion. The conversion of the steel
was less rapid, the ingots smaller, and liquation was slight. The
present conversion, in larger vessels, teemed in larger ingots,
are not as sound, for the entire length, as the smaller ingots.
The tendency of the upper portion is to form a pipe or become
porous, or from the longer time in cooling, allows the metalloids
to separate from the bath and become concentrated in an upper
central core of the ingot. This does not produce homogeneous
metal in the head. It is not capable of sustaining the wheel
contact pressures by its cubic elasticity, and fails rapidly in
service.
These are important problems in the manufacture of rails
for the present service. It is not alone a question of heat treat-
ment, a necessary, yet an over-estimated, panacea for defective
steel, but to secure a sound ingot, so that the metal in the entire
head will be homogeneous.
The steel in the rail section should be sound and have suffi-
cient physical properties and rigidity of structure to preserve
the shape of the section under the traffic, except the loss by wear
of the wheel treads in the bearing surface, and the wheel flanges
on the side of the head. Sound metal of 56,000 or more granu-
lations per square inch, and elastic limit of 56,000 to 60,000
pounds, has sustained the present wheel loads without distortion
of the heads. The facing ends of the rails should not flatten in
sections having moments of inertia of more than twenty-five
fourth-power inches. More limber sections are not finished
sufficiently secure to prevent the wear of the facing ends of the
rails under the shocks of the present wheel loads.
I have under observation rails made at several different
mills; and in those without effort to control or check liquation,
the splitting and the piping of the rail heads is pronounced.
154 The Iron and Steel Magazine
In rails rolled from metal where attention was paid to checking
liquation, the piping of the rail heads is practically unknown.
The problems of checking liquation in large ingots are not always
easy of solution and must be solved in reference to the practice
of each mill.
The metal for the stiff rails, the efficient engineering struc-
tures, must be sound for any section to stand in the track with-
out distortion imder the traffic. If the steel contains a central
core of harder material than the outside, or is porous, containing
occluded gases, then it does not have sufficient toughness and
tenacity of structure for the requisite limits of cubic elasticity
to sustain the wheel load effects without distortion, and fails
mechanically as an engineering strticture.
HARD CAST IRON: A THEORY OF ONE OF ITS CAUSES *
By HENRY SOUTHER
\ CONSULTING metallurgical engineer in contact with
-^"^ machine shops is accustomed to hearing comxplaints from
the operators of machine tools of hardness of material being ma-
chined. Sometimes it develops, especially with steel, that
the trouble is not that the steel is hard; but, on the contrary,
that it is exceedingl}^ soft. The softness is of such a character
that the edged tool does not succeed in cutting the steel keenly,
but rather tears it off, some of the particles clinging to the edge
of the tool, causing excessive friction and rubbing, and drawing
the temper of the tool, and dullness soon follows. The effect, as
far as the machine operator is concerned, is that of a hard steel;
the tool is spoiled. This not only applies to low carbons but to
a peculiar physical condition of higher carbons, say, in the neigh-
borhood of .50, due to bad annealing.
Then there is the legitimate hard steel, which is really hard
in the true sense of the term.
Cast iron that chills may be called hard in the trtiest sense
of the word. That is the complaint that is most often met when
the term hardness is used in connection with machining cast
iron. Iron chills because of high sulphur or low silicon, or a
* Presented at the June, 1Q05, meeting of the American Society for
Testing Materials.
Hard (\ist Iron 155
combination of both, and machine tools simply cannot cut it.
This kind of hardness is more often found in thin work than in
thick work. For example, the hardware people casting very
thin material have to use the softest of iron, high silicon and low
sulphur, in order that the small amount of machining they do
may be done at all.
In the last five or six years three separate complaints of hard
iron have reached the writer and proved of so baffling a character
that in each case visits were made to the machine shops working
the iron and the complaint carefully investigated. Analysis
or test did not reveal the cause.
The most instructive case covers them all. This instance
was most instructive because it occurred on a multiple drill
where several different sizes of standard drills were used and
several thicknesses of metal w^ere involved. On approaching
the machine it was noticeable at once that there was trouble,
because the drills were screeching in an unusual way.
It developed that small drills, J inch or thereabouts, were
standing up with this iron just as well as any other, but the
larger drills in the neighborhood of h inch and f inch were
dulling exactly as though the iron were charged with emery.
The edges were being ground off and would only last a small
fraction of the time usual for the same drills in the same
machine.
Here was an unusual condition, — thin iron working easily;
thick iron on the same castings working with difficulty.
The chemical results w^ere normal, except manganese:
silicon 2.50, phosphorus .70, sulphur about .080, total carbon
3.50 and manganese .16.
The fracture of the iron was good, and moreover it was
quite normal as -far as could be seen with eye or microscope.
A Keep's test drill was used, and developed nothing unusual,
no signs of hardness, thick and thin iron showing a normal curve.
There Vv^as no opportunity to test the tool-wearing qualities
on the Keep machine, because the drill was sharpened after
every hole drilled.
Inasmuch as the only abnormal part in the analysis was
shown in the manganese, that element was suspected, although
there seemed to be no metallurgical reason for so doing. Means
were taken to raise it to the neighborhood of .50, and as soon as
1^6 The Iron and Steel Magazine
this was done the difficuUy disappeared in the machine shop
and has not reappeared after some months.
It is a com.plaint which reached me, as I said above, from
two other sources, and in both of these sources the complaint
was described by the machine shop people by saying that the
iron was gritty. They were fully convinced that there was sand
in it, but examination showed that to be out of the question and
the iron was as clean as any iron.
This leads me to believe that there must be some carbide
of iron or carbide of silicon that forms in the absence of a reason-
able amount of manganese, and that does not form with man-
ganese present. What this chemical combination may be I
cannot surmise, but the problem presented is an interesting
one from a theoretical and practical standpoint. Apparently
its cure has been found, but the question remains, — Why?
The actual castings causing this trouble were put through
and no specim>ens kept, so that the writer can furnish po samples
for study, but this is doubtless something tliat comes before the
other members of the Society; and although specimens may not
be easy to get, they can doubtless be foimd.
ETCHING OF HIGH CARBON STEEL *
By E. H. SANITER
TTAVING experienced considerable diffictdty in getting etch-
-*■ ^ ings of high carbon steels, especially if in the tempered
condition, with iodine, 2 per cent nitric acid or picric acid, I tried
Sauveur's method of dipping in strong nitric acid (T.42 Sp. G.)
and washing at the tap. This gave better results but required
several treatments to get the desired etching. I then tried dip-
ping the specimen in absolute alcohol, then in strong nitric acid
and then washing at the tap. This gave a very good etching
with only one treatment. The precaution must be taken to
move the specimen, about in the acid rapidly to get an even
attack; to do this I hold the specimen in a pair of forceps. It
is also necessary to use fresh nitric acid for each etching.
♦Received July 17, 1Q05.
ABSTRACTS *
■ {From recent articles of interest to the Iron and Steel Metallurgist)
np HE Application of Dry-Air Blast to the Manufacture of Iron.
-*- Joseph W . Richards. A discussion of the paper of Mr.
Gayley read by title at the Lake Superior meeting of the Ameri-
can Institute of Mining Engineers. 8,000 w. — The following
extracts clearly indicate the author's views regarding the causes
of the great saving of fuel resulting from the drying of the blast.
" The whole qtiestion of the economy obtained converges
towards the discussion of the generation of the heat necessary
for smelting in the region of the tuyeres. To generate this
amount of heat for the requisite temperature, with the produc-
tion of no more carbon monoxide than is necessar)^ to achieve
reduction of the charges above, is the direction in which economy
is to be obtained. The efficiency of hot blast is due to exactly
this reason, that it increases the smelting-power of the tuyere
region without an increase in the carbon monoxide formed, and
with a higher temperature of the gases in the tuyere region.
" In another view of the matter, furnaces are limited by lack
of smelting power at the region of the tuyeres rather than of
reducing power in the upper part of the furnace, and any pro-
cess which increases the former, without diminishing the latter
below certain requisite limits, will strengthen the furnace at its
weakest point and result in corresponding economy. The hot
blast accomplishes this by the positive addition of heat bodily ;
the dr}^ blast improves matters by saving some heat otherwise
* Note. The publishers will endeavor to supply upon request the full
text of the articles here abstracted, together with all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60
cents, "D" 80 cents, "E" $1.00, "F" $1.20, "G" $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cases
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
ing upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.
157
15^^ The Iron and Steel Magazine
lost in decomposing the moisture, thus also increasing the smelt-
ing power. The drying of the blast is the exact equivalent
of extra heating of the blast, both in its immediate and in its
ultimate effects, but with the additional benefit of increasing
the regularity of the temperature, smelting power and general
running of the furnace.
"The figures show an apparent increase of only 4.8 per cent
in the efficiency of utilization of the heat generated at the tuyeres
for smelting purposes, but this is not the correct way to regard
the matter. In the two cases quoted there is an equal smelt-
ing done for the burning of 75.3 kilograms of carbon at the
tuyeres in the one case and of 5 8. 05 kilograms in the other, a
clear saving of 17.25 kilograms of carbon burned at the tuyeres
per unit of pig iron and slag melted, or 22.9 per cent of the carbon
consumed at the tuyeres. This represents approximately,
therefore, the total economy in fuel for the whole furnace. More
important than that, however, it also rej^resents increased
speed of smelting, by the use of a constant blast. A constant
blast burns a constant amount of carbon at the tuyeres in a given
time, but if the furnace can be made to smelt a unit of pig iron
and slag with 22.9 per cent less of carbon burned at the tuyeres,
it should be expected, by using a constant blast, that the rate
of smelting would be increased at least 22.9 per cent by the new
conditions, provided, of course, that the quantity of blast,
measured at standard conditions, was kept constant.
"A more logical way of viewing the matter of increased
smelting capacity is to consider the conditions with the use of
moist blast, and then to consider what would happen if the moist-
ure were suddenly to disappear from the blast, and the 14,512
calories absorbed in decomposing it to be suddenly restored to
the furnace. The result would undoubtedly be a stidden increase
in the heat available, and a rise in temperature of the melted
iron and slag. Suppose this rise of temperature of the products
were counteracted by increasing the burden of the furnace to
such an extent as to bring back the temperature of pig iron and
slag to their primary state. When this condition of extra bur-
dening had been reached, the temperature of the pig iron and
slag would be the same as before removing the moisture, and
the 14,5:2 calories would have been all absorbed in doing an
increased amount of smelting. In the actual case, the 14,512
Abstracts 15Q
calories were not all saved, but 14.512 — 3,225 = 11,287 calories
were, by removing 77.7 per cent of the moisture; and the burden
was correspondingly increased to keep the quality of pig iron
made the same as at first.
" I conclude, that since Mr. Gay ley obtained an increase of
24.86 per cent, the views here advanced give a satisfactory
explanation of the increased smelting power of the furnace,
that the heat saved by absence of moisture is directly utilized
for increased smelting capacity, aided furthermore by a higher
heat jjotential which of itself alone would increase the smelt-
ing rate about 5 per cent, the two factors working simultaneously
and necessarily interdependently towards the total effect.
" In conclusion it may be said that the increased efficiency
obtained by Mr. Gay ley could theoretically have been obtained
by an increased temperature of blast alone, viz., by using the
moist blast at 597° C. (1,107° F.), instead of at 382° C. (720° F.).
Such an increase would produce the effects of quicker running and
economizing coke to the quantity noted with dry blast, but
would still leave the furnace subject to the irregularities insep-
arable from using ordinary air with its varying temperature
and content of moisture. The increased regularity of running
of the furnace and quality of product, due to uniform tempera-
ture of air supplied to the blowing engine and uniform tempera-
ture before the tuyeres, is the fundamental economic justification
for Mr. Gay ley's innovation; the increased rate of driving and
economy of fuel alone could be obtained more cheaply by in-
creasing the capacity of the stoves." No. 396.
The Manufacture of Chilled Wheels. P. H. Grifhn. " The
Railroad Gazette," June 16, 1905. 4,000 w. — The following
comments are extracted from this paper:
" The chilled wheel owes its origin to the peculiar high chilling
property of iron smelted with charcoal in the blast furnace,
which property is not possessed by iron smelted with coke. Of
late years the use of non-charcoal irons and scrap has become
quite general, and many chilled wheels have been put in service
which have contained large percentages of such material. The
use of steel scrap to harden, and ferro-m.anganese to toughen
and give chilling properties, has established the use of such
non-chilling irons, but the quality of chilled wheels has not been
i6o The Iron and Steel Magazine
improved by such practice. The indiscriminate delivery of
scrap wheels in part payment for new wheels, and the conse-
c[uent use of such scrap in making new wheels has gradually
distributed the non-charcoal iron wheels throughout all wheels,
and thtis spread the undesirable metal in all directions. It
must be remembered that every chilled wheel made is destined
to a sort of reincarnation, so that it will probably go on to the
end of time bearing in each successive life the burden not only
of its own defects, but of those acquired from its fellows in the
foundry cupola. The manufacture of charcoal pig iron is one
of the oldest industries in this country. It has increased from
year to year, particularly in the Lake vSuperior region. The
present annual capacity of all charcoal furnaces in America
exceeds 500,000 tons. The supply is, therefore, ample so far as
requirements for chilled car wheels are concerned.
" The wheel made some years ago may have had its failings,
but it certainly was made of better material than the average
wheel of to-day, which is bound to progress toward a lower
average quality of material, on account of the general practice
of recasting old wheels into new ones, and the increasing use
of non-chilling or non-charcoal iron and scrap. Railroads,
should buy, at least for special service, such as 50-ton car equip-
ment, wheels made from new charcoal iron. It is not necessary
to specify mixture or kinds of iron, but simply to provide for
wheels of new charcoal iron. The additional cost will not be
great. In the case of wheels bought at average prices the added
cost should not be over eight to ten dollars per car.
" The possibilities of the chilled wheel for meeting conditions
of service that are likely to arise in the future, lie in two direc-
tions. First, increase of strength and resistance to wear in the
material of which the wheels are made and improvements in
type and section of wheel. Second, improvements in methods
of manufacture.
"The principal factors of value in charcoal iron for chilled
wheel manufacture are strength in general, and hardness com-
bined with wearing quality in the chilled surface obtained. By
the use of special alloys some remarkable qualities of chilled
iron have been produced. In the course of some recent experi-
ments made by the writer, chilled metal was produced in which
the crystalline structure was composed of what are known as
Abstracts i6i
hair crystals. These crystals have no angles but interweave
with and radiate from each other in a complex but perfectly
ordered arrangement. The strength of this chilled metal is
equal to that of the best crucible steel, and the increased strength
is due to the process of chilling. This reverses the ordinary
result when chilling occurs, which is, that brittleness is increased.
" There is not the slightest doubt that chilled wheels equal
to and beyond any of the present demands of service can be
readilv obtained by railroads that are willing to pay for them.
No more striking proof of what is practicable and indeed neces-
sary in this direction can be afforded than by consideration of
the fact that the entire net cost of wheels for a car is less than
$25. Eight chilled wheels, t^t, inches in diameter, weighing 600
to 700 pounds, less scrap value, which can always be realized
at 50 per cent of first cost, will not exceed the net cost stated.
This is less than 5 per cent of the average cost of the new car.
"The chilled wheel m.akers have certainly done their part
in reducing the cost of building and operating American rail-
roads, and there must be some limit to the further demands on
them to produce vStill better results regardless of conditions." No.
397. B.
Power from Waste Furnace Gases. " Power," June, 1905,
3,800 w., illustrated. — The article describes briefl}^ the in-
stallation of the Koerting two-cycle gas engines at the Lacka-
v/anna Steel Company's plant at Buffalo, N. Y., as well as some
European plants. It concludes as follows:
" The power available from the waste gases in a blast-furnace
plant is known to be more than enough to generate electricity
for supplying the motors for all purposes about the power plant,
and this especially so if coke-oven gas is available from works
which are part of the same plant. In this case, the excess power
can be used for the manufacture of many by-products, such as
carborundum, calcium carbide, etc., as is being done in several
European works.
" It is possible to produce a very high grade of steel by means
of the electric furnace, and there is little doubt but that electric
furnaces will be utilized in the near future in many plants in
this country and abroad for the production of a high-grade steel.
Steel is now being made in Sweden, Switzerland, Erance and
1 62 The Iron and Steel Magazine
other European countries by electric fusion with great success,
and another appUcation for electric current in iron and steel
plants worthy of consideration is that of the electromagnetic
concentration of iron ores, which has been applied with greatest
success in this country.
" Where the electric power generated in these plants cannot
be all utilized to advantage in or about the works it is suggested
that by means of suitable transmission lines the surplus power
may be utilized for operating electric railways and also for light-
ing the streets of adjacent cities. The reventie from such a
source would be a considerable amount, and in a sense would
almost make the operation of the blast furnace for the production
of ore an extremely small item, since the furnace itself is indeed
a very efficient gas producer." No. 398. B.
Effect of Manganese in Low Silicon Cast Iron. H. C. Loud-
enbeck. Paper read before the American Foundrymen's Asso-
ciation, New^ York, June, 1905. 1,800 w. — The author reports
the results of some tests made with a view of ascertaining
the effect of manganese on the chill and fracture of cast iron
having a low percentage of silicon, and from these he concludes
as follows:
^' Manganese can be used to advantage in low silicon and
chilling iron in the following cases:
" In mixtures where the percentage of scrap is large and the
sulphur necessarih^ high (this will occur in a car-wheel mixture
where usually a large portion of old metal is used) the result of
this increase in manganese would be lower sulphur, lower com-
bined carbon, less chill and greater strength.
'' Very often chilled plates are required having hard chilled
faces and soft' backs suitable for planing. Manganese added in
the right proportion will reduce the tendency to mottle and
make a comparatively soft graphitic back.
'' In all cases where chilling irons are melted in a cupola and
the sulphur is over .7 per cent the iron can be strengthened by
the use of ferro-manganese or pig iron having a high percentage
of manganese.
'^ There are some cases where the manganese should be kept
low. In the manufacture of large hydraulic cylinders it is neces-
sarv to have a close mottled iron to withstand the pressure and
Abstracts 163
prevent leakage. If the nianganese is too high this mottled
structure is replaced by a coarse graphitic structure, which is
not satisfactory for this class of work." No. 399.
Iron Ores Briquettes for the Blast Furnace. Henry Louis.
*' Gassier 's Magazine," JuU^ 1905. 6,500 w.. illustrated. — The
author's views regarding the briquetting of iron ores are summed
up in the following words :
*' It is obvious that the briquetting of iron ores has long
since passed the experimental stage, and that the smelting of
briquettes in the blast furnace is also a proved economic and
practical success. There is no reason to dotibt that this method
of dealing with finely divided iron ores, whether artificially or
naturally comminuted, will be further extended in the near
future. By its means such coniparatively low-priced material
as purple ore or pyrites residues can be converted into high-class
iron ores, and seeing that Great Britain imports annually about
750,000 tons of pyrites, this one item alone is of considerable
importance. Furthermore, cheap methods of crushing and
concentration, combined with briqtietting, will make it possible
to smelt with advantage many an iron ore too poor to treat directly
in the blast furnace, and the future alone will show how far the
mechanical elimination of impurities from iron ores can be carried
profitably. Technically, the problem is solved, but its economic
limiits yet remain to be defined." No. 400. B.
Water-Cooled Ports for Open-Hearth Furnaces. " Iron Age,"
May 4, 1905. 1,300 w., illustrated. — The article describes a
system of water-cooled ports for open-hearth furnaces installed
at the Illinois Steel Company plant and at some other plants.
The device is the joint invention of Geo. L. Davidson and David
R. Mathias. No. 401. B.
Microstructure and Frictional Characteristics in Bearing
Metals. Melvin Price. Paper read before the American Society
of Mechanical Engineers, Scranton, Pa., June, 1905. 10,000 w.,
illustrated. — The author reports the results of an investigation
carried out in the laboratories of Columbia University. No. 402.
The Transfer of Heat at High Temperatures. Frank C.
Wagner. Paper read at the Scranton meeting of the American
164 The Iron and Steel Magazine
Society of Mechanical Engineers, June, 1905. 4,000 w. — The
primary object of the experiments described in this paper was to
determine the time required to raise plates of iron and steel to
a welding temperature in an open-hearth regenerative furnace-
No. 403.
The Cupola System of the Michigan Stove Company. W.J.
Keep. A paper read before the American Foundrymen's Asso-
ciation, New York, June, 1Q05. 2,100 w. No. 404.
Cast Iron. Crushing Loads and Microstructure. W. J.
Keep. A paper read before the American Society of Mechani-
cal Engineers, Scranton, Pa., June, 1905. 5,000 w.,47photo-
Hiicrographs. No. 405.
METALLURGICAL NOTES AND COMMENTS
Paul Louis Toussaint Heroult (see frontis-
oms oua- p^g(.g\ ^,g^g born at Thury-Harcourt (Calvados,
samt Heroult r / j .. .
France), April lo, 1863. After attending the
preparatory school attached to the vSchool of Mines, Mr. Heroult
in 1884 entered the employ of an industrial firm, before complet-
ing his engineering studies, and began at once his remarkable
investigations dealing with the extraction of aluminum. In
1886 Mr. Heroult obtained a patent for his process of producing
aluminum by an electrolytic method, which was destined to
completely revolutionize the metallurgy of this metal, reducing
its price from 100 francs in 1884 to 2.5 francs per kilogram in
1 90 1. In 1887 Mr. Heroult obtained a patent for what may be
considered as the first really industrial electric furnace. Our
readers are too familiar with Mr. Heroult's recent development
of the electric furnace applied to the direct extraction of iron and
steel as well as to the refining of cast iron to require more than
a passing notice here. As stated elsewhere in this issue Mr. He-
roult is to take charge of the experiments to be conducted by the
Canadian Government to ascertain the practical value of electric
smelting under the conditions prevailing in that country. Mr.
Heroult's results will be awaited with much interest by iron
metallurgists of all countries.
Last January Mr. Heroult was awarded the Lavoisier medal
by the Societe d' Encouragement pour 1' Industrie Nationale.
The eighth annual meeting of the American
,. . -_ . , Societv for Testing: Materials was held at
iesting Materials - ^
Atlantic City, N. J., from June 29 to July i,
some 200 members being in attendance. The secretary reported
an increase in membership of 192, bringing the total membership
to 677, while the number of technical committees has been in-
creased from II to 18. W. H. Bostwick and John McLeod
were elected to fill vacancies in the executive committee. Re-
165
i66
The Iron and Steel Magazine
ports of the various committees were presented at the meetings
and the following papers of special interest to our readers were
read and discussed :
'^ Protection of Iron and
Steel Structures by Means of
Paper and Paint," by Louis
H. Barker.
'^ Some Causes of Failure
of Rails in Service," bv Robert
Job.
'' Influence of Methods of
Piling Staybolt Iron on Vibra-
tory Tests," by H. V. Wille.
*' A Preliminary Report
on Tests of Nickel Steel and
Carbon Steel under Combined
Stresses," by E. L. Hancock.
"A Comparison of Stand-
ard Methods of Testing Cast
Iron," by Richard Moldenke.
C. B. Dudley, President American
Society for Testing Materials
" Hard Cast Iron: The
Theory of One of its Causes,"
by Henry Souther.
'' The Thermit Process
in American Practice," by
E. Stuetz.
^^ Rail Sections as Engi-
neering Structures," by P. H.
Dudley.
Three of these papers will
be found reproduced in full
in the present issue of The
Iron and Steel M agazine ,Yfh.i\e
the others will be printed in
subsequent numbers.
The Thursday evening
session was held jointly with
the Society for the Promotion
of EngineeringTEducation, and President Dudley read an excel-
lent address on ''The Testing Engineer."
Edgar Marburg Secretary-Treasurer
American Society for Testing
Materials
^[ctaIIl(ri^icaI Ahtcs and Comments
167
The American Foundrymen's Associa-
American Foundrymen's ^. 1 1 1 ., • 1 - 1 -• • tvt
. . ^ tion held its eii^hth convention m New
Association. ^
York, June 6 to 8. The usual reports
of committees were presented and many papers read and dis-
cussed, amoniT which we quote the following as of special interest
to our readers :
" Notes on Some Retort Coke Melting Ratios," by C. M.
Schwein, Milwaukee, Wis.
" Melting Steel with Cast
Iron," by R. P. Cunningham,
Holyoke, Mass.
'' Notes on Pipe Foun-
dries and Suggestions on Metal
Mixers for Foundry Pur-
poses," by J. B. Nau, New
York City.
''Some Thoughts on
Modern American Foundry
Practice," by John C. Burns,
Plainfield, N. J.
''The Effect of Manga-
nese in Low Silicon Cast
Iron," by H. C. Loudenbeck,
Wilmerding, Pa.
'' Blowers, Piping and Cupolas at the Plant of the
Michigan Stove Co.," by W. J. Keep, Detroit, Mich.
'' The Use of Thermit in a Railroad Shop," by Jas. F. Webb,
of Elkhart, Ind.
'' Report of the Committee on Standard Methods of Deter-
mining the Constituents of Cast Iron."
'' Fan Blower v. Positive Pressure Blower," by H. F. Field,
Pittsburg, Pa.
'' A Practical Foundry School," by W. C. Bruce, Cleveland,
Ohio.
Some of these papers will be found abstracted in the present
issue of The Iron and Steel Magazine, while the others will be
puVjlished or reviewed in subsequent issues.
Thos. D. West, President-elect Ameri-
can Foundrymen's Association
The Carnegie Steel Cross-Tie. — The Carnegie Steel Com-
pany, Pittsburg, Pa., has for some time been furnishing promi-
i68
The Iron and Steel Magazine
nent railroad companies with considerable quantities of steel
cross-ties. It is believed that the near future will witness mate-
rial development in the use of metal ties to replace the wooden
ones. Wooden ties are steadity growing scarcer and dearer.
Herewith illustrations are given of the new tie which has been
brought out by the Carnegie Steel Company, together with the
method of its use. Fig. i is a cross section of the tie, which,
it will be observed, is an exaggerated form of the well-known
I-beam. This section combines ample bearing surface of a proper
shape for bedding and tamping, sufficient surface for seating
the rail, the greatest rigidity and transverse strength for a given
weight of material, and an easy means for securing the rail to
Fig. I . Cross Section of the Carnegie Steel Cross Tie
the tie. It has a top flange 4^ inches wide, a bottom flange 8
inches wide, a depth of 5^ inches, is 8 feet, 6 inches in length,
and weighs 19.7 pounds per foot , or a total weight of 167.4 pounds
per tie, exclusive of fastenings, which weigh about 6 pounds.
The rail is secured to the tie, as shown in Fig. 2, with four three-
quarter-inch bolts, by means of rolled steel clips, fitting accu-
rately on the flange of the rail. These clips have a bevel exactly
the same as that of the flange of the rail and are carefully punched
so that the shoulder of the clip gives proper and positive rail
alignment. The necessary insulation, where automatic block
signals are in use, is provided for by the use of wooden shims
Metallurgical Notes and Coiiuiiciits
169
between the rail and the tie, liber bushings around the bolts and
tiber washers under the nuts.
These ties have been used in svifficient quantities to demon-
strate their efficiency during the past two years on lines carrying
fast and heavy traffic, as follows : Duluth & Iron Range Railroad,
Bessemer & Lake Erie Railroad, New York Central & Hudson
River Railroad, Lake Shore & Michigan Southern Railroad.
An order has just been placed by the Bessemer & Lake Erie
Railroad with the Carnegie Steel Company for ten miles of
these ties, the order amounting to about 2,100 tons. This order
follows a test of a half mile of the Carnegie steel tie for the past
Fig. 2. The Carnegie Steel Cross-Tie as applied at a joint
six months. Arrangements are being made to have the tie used
on the Pennsylvania Railroad, near Emsworth, Pa. '' The
Iron Age," June 22, 1905.
The Electro-Metallurgy of Iron Alloys. — Although much
has been said about the practicability of using the electric
furnace for the smelting of iron and the refining of steel, it
has been realized that the element of cost must, at the
present time, prevent such methods from entering into general
competition with the older processes. At the same time there
is a field in which the electric furnace can be employed where
lyo The Iron and Steel Magazine
the question of the cost of the product is secondary to its
quality, and for such work the electric furnace possesses numer-
ous advantages. In recent issues of " Stahl und Eisen " there
is given an account by Herr V. Englehardt, of the progress which
has been made in the practical operation of the Kjellin process
for producing high-grade alloy steels in the electric furnace, and
some abstract of these papers wiU be found of interest.
The experimental researches of Kjellin have been conducted
since 1899 at the Gy singe Iron Works, situated on the Dalelf,
about four hours by rail from Stockholm in Sweden. These
works operated a blast-furnace forge and sulphite-cellulose
works, this latter being the development of the facilities of the
neighboring forest, and here in March, 1900, the first electric
furnace, with a capacity for utilizing 78 kilowatts of electric
energy, was put into active operation, turning out 270 kilograms
of cast steel in twenty -four hours, with a consumption of about
7,000 kilowatt-hours of electric energy per metric ton of steel.
This was soon improved by the production, in a second furnace,
of 600 to 700 kilograms of steel in twenty -four hours, with a
consumption of 2,140 kilowatt-hours per ton. In July, the
cellulose works were destroyed by fire, and the energy that had
been required for this portion of the establishment was diverted
to the service of a third and still larger furnace, in which still
more successful results have been obtained.
The Kjellin process is practically a reproduction of the
crucible process of making steel by fusing the materials in such
proportion as to give the required carbon content to the product
without requiring any subsequent additions to the charge. The
only differences are the substitution of electric methods of pro-
ducing the required temperature and the arrangement by which
larger quantities can be handled than is practicable with the
crucible process. Further, the Kjellin furnace is an induction
apparatus, there being no contact of electrodes or any external
parts with the charge, and hence there is no possibility for the
introduction of any impurities, either solid or gaseous, so that
the purity of the product is dependent entirely upon the purity
of the material with which the furnace is charged.
The furnace itself is practically a stepdown transformer,
and the simplicity of its construction is marked. In an ordinary
transformer the high-tension current passes through a coil of
Mctallin'i:^{cal Xotcs ami C\vii}nciits 171
proper fineness, wcund about an iron core, while around this
again is wound the coil of heavier wire in which the induced
current is produced when an alternating current traverses the
inner coil. In the case of the Kjellin furnace the outer coil is
replaced by an annular channel, or sort of gutter, formed in the
refractory material of the furnace, the wire coil and its iron
core being in a central space so arranged that a free circulation
of air is permitted. The charge of metal in the annular trough
or gutter forms the short-circuited outer winding, so to speak,
of the transformer, and the heat generated in the mass of the
metal is employed to effect the fusion of the combination. It
is evident that the capacity of the trough may be made sufficient
to hold the charge required, and that this ring-like crucible may
be kept closely covered during the entire operation, the completed
run of steel being drawn off by a tap hole into a ladle and poured
into molds as desired.
The latest furnace of this type differs from the earlier ones
in dimensions only, the simplicity and effectiveness of design
having been such as to render no radical changes or modifications
necessary.
From the records of the works it appears that the output
of a furnace is about 5,000 kilograms of steel in twenty-four
hours, the power required being 167. i kilowatts, or about 222
electrical horse-power, giving 4,010 kilowatt-hours per day, or
802 kilowatt-hours per metric ton of steel. Records for another
day show a better result, the consumption of electrical energy
being 770 kilowatt-hours per ton.
Herr Englehardt gives a very complete thermal analysis
of the working of the furnace, showing* that the theoretical
consumption of energy should be 489 kilowatt-hours per ton
of steel, so that the above performances indicate an efficiency
of more than 60 per cent for the actual process.
The current at Gysinge is at present derived from water
power, and under these conditions it is maintained by Kjellin
that the highest grade of crucible steel can be made at a lower
cost than by the crucible process. It is suggested also that
such furnaces may well be operated in connection with blast
furnaces, the electrical energy being obtained from the waste
gases of the blast furnaces, used in gas engines, thus furnishing
a method of utilizing the power in the gas in an allied line of
1^2 The Iron and Steel Magazine
work instead of using it to generate electric current to supply
an outside market.
The Kj ellin process has also been used to work with pig and
ore as well as with pig and scrap, and the results have been
encouraging when a high grade of ore is employed.
It is evident that the economy of the process is dependent
almost entirely upon the cost of the electrical energy, and com-
putations for one locality cannot therefore be used for another
unless the comparative cost of current is known. Data for a
large furnace, to consume i,ooo h. p., are given for a daily pro-
duction of 30 to 36 tons, from which it is computed that steel
can be made for about 70 marks per ton, but practical operations
have not yet been carried out on such a large scale. '', Engi-
neering Magazine." May, 1905.
The Metallurgist and His Work. — The man to whom this
title applies has to control vast operations of a widely varied
nature. So far, only Kipling has immortalized the engineer in
that poetical contribution to technics, '' M'Andrew's Hymn."
Some day, we trust, he will render equal justice to the metallur-
gist, for, indeed, without the latter the civilized world would
revert to barbarism. The engineer, the electrician, the physicist
and the chemist may all plan, but it is eventually the metallur-
gist upon whom they rely. To repeat again our first president's
remark, '' There is hardly a necessity or convenience of life to
the production of which iron has not contributed."
To our brothers in the technical professions we freely admit
our indebtedness, but the jons et origo is the metallurgist, who
enables the ideas of the engineer and the physicist to be carried
into effect.
The three main arteries of modern progress — the railway,
the steamship and the electric telegraph — would never have
existed without the aid of metallurgy ; there are still many great
and complicated problems in that field to be solved. Higher
speeds, increased wear and tear, the outcome of modern require-
ments, demand the consideration of most complex questions;
but I would ask, '' Have we not risen to the occasion? " Do we
not guard well the lives and safety of the people ? On the steel
rail and on the steel ship-plate, how entirely the civilized world
depends for its communications, nay, for its existence! In the
Metallurgical Notes and Comments 173
science of war the i?-inch gun hurls its projectile with a muzzle
velocity of about 2,600 foot-seconds, and with the enormous
energy of 42,000 foot-tons stored in it. How mighty are the
powers of resistance to stress in the steel of which it is constructed !
The recent development of high-speed cutting steel is revolu-
tionizing machine-shop practice throughout the world. From
the iron-manganese alloy, containing 4 per cent to 5 per cent
of manganese, and but little over .40 per cent carbon, which in
its cast state is so friable that it can almost be powdered between
the fingers, to chromium steel having an elastic limit of 70 tons
per square inch, and 100 tons or more tenacity, or to the nickel-
manganese alloy, with a tenacity of 60 to 65 tons, which draws
out cold in the testing machine to 60 per cent or 80 per cent of
its length, is a wide range of production; and there are a thou-
sand and one types and varieties between these limits.
It has been well said that metallurgy, like applied chemistry,
appears to be comparatively simple in its practical aspect; but
theoretically it is extremely complex and varied.
The only possible way to make satisfactory progress is to
combine the practical with the theoretical in the training of
our metallurgists. Rule-of -thumb methods no longer suffice.
Much of the mystery which has surrounded our science has been
swept away; and rightly, for it is only by the intelligent appli-
cation of scientific principles, balanced by and co-operating
with sound practical knowledge, that the metallurgist can meet
the complex requirements of the day. He is neither completely
chemist nor completely physicist, but he owes a debt of grati-
tude to both. Many are the problems submitted for his considera-
tion. He is brought in contact with the needs and difficulties
of engineers of all branches. Whilst the naval architect and the
military engineer demand from him, on the one hand, armor
impervious to penetration, the artillerist, on the other hand,
asks for projectiles which shall, without showing crack or blem-
ish, pierce this impenetrable armor.
In speaking of the importance of the metallurgist, it can
be safely said that, except for his labors, the world would never
have emerged from the darkness of the Middle Ages ; if we were
without iron there is hardly a single modern advance which
would not be absolutely and irretrievably lost.
As our first president pointed out in his opening address,
174
The Iron and Steel Mamzine
'' It is hardlv possible to name a single necessity or convenience
of life to the production of which iron has not contributed."
This applies now with even more force than in 1869, when it was
uttered. From R. A. Hadfield's Presidential Address, '' Iron
and Steel Institute," May, 1905, meeting.
The Position of the Steel Foundry. — The steel casting
industry in the United States has had not a little attention in
recent years. Manufacturers of gray iron and malleable castings
have been hearing from time to time of the substitution of steel for
their products, and there has been also some use of steel castings
instead of forgings. However, an examination of the sta-
tistics of the industry would indicate that perhaps undue impor-
tance has been attached to the effect of the displacement of gray
iron and malleable castings by steel castings. It is true that
cast steel frames for locomotives are taking the place of forged
frames, and that this means 5 tons of steel castings for the aver-
age locomotive. It is true, also, that draw bars are now largely
specified to be steel castings instead of malleable castings, and
that car bolsters and center plates and wheel centers come from
the steel foundry. It is well known, moreover, that steel rolls
and housings and other steel castings are increasingly in use in
rolling mills. In the discussions on the inroad of steel upon gray
iron these facts have been given prominence, and the deduction
has been made that gray iron and malleable castings have a
steadih^ narrowing field.
The statistics of the American Iron and Steel Association
show that the production in the past five years of the various
classes of steel castings has been as follows :
PRODUCTION OF STEEL CASTINGS IN THE UNITED STASES. GROSS TONS
iQoo 1901 I go 2 1903 T904
Acid open-hearth ... 134,847 206,681 255,475 265,469 203,915
Basic open-hearth 42,644 94,941 112,404 134,879 98,919
Bessemer 6,467 6,764 12,548 18,059 1^,051
Crucible 3,989 3,927 4,955 5,409 4,3oS
Miscellaneous 4,856 5,257 5,553 6,409 7,018
Totals 192,803 317.570 390,935 430,265 330,211
It will be seen that the falling off in 1904 from the high
record reached in 1903 was marked, being no less than 23 per
Mi-Uilliire.iiiil -Vc/.-.T .nni Comments i75
cent and that the output of steel castings last year was but
"ttle more than m ,,0,. There are no statistics of the produc-
t?on of c^rav iron, chilled iron and malleable castings, but some
ompari';on can be made between the two years by taking the
production of foundry and malleable Bessemer iron. For 1903
the production of foundry iron m the United States -s 4^409 o ^3
aross tons and of malleable Bessem.er 473.781 tons. Last year
the lodu'tion of the two classes of iron amounted to 3,8^.^9
toL'and .63,5^9 tons, respectively. This -^-^e- dechne
of 16.2 per cent, but stocks of iron were l^f °«.^^^;5^^^\^^
than on January i, 1904, so that the actual fallmg off m the
pr duc^ii of grly uon and malleable castings last year as com
pared with 1 9°3 was probably less than 1 5 per -^^^The greater
shrinkage in the steel castings trade was largely due, doubtless
to the failure of railroad demand last year. Another noteworthy
eature of the statistics is the relatively small tonnage of Bes-
emrsteel castings produced m the United States^ So -ch
has been said about the rapid introduction of the small Bessemer
converter !n steel foundries that it might be expected to have
made a larger impression upon the industry. But as the small
TnTerters are employed, as a rule, on small castmgs calling for
Wh heat and fluidity m the metal, tonnage does -t count up
rSdly The average for 1900 and 1901 was nearly doubled n
rfwith an output of 12,548 tons, then increased about 50
;:rcent, to 18,099, m 1903, and fell back to 16,051 tons m
'''°'But perhaps the most striking impression that will be ob-
tained from the figures is that of the relatively small proportion
of the total of castings produced m the United States that comes
fom steel foundries It is probable that not more than 6 per
cenTof the total of iron and steel cast m sand molds m X904 was
"'""\n recent months the steel casting industry has shown a
good recovery from the stagnation that especially marked 1 m
!oo4 From the 2-cent level, touched at the worst of last
a -s depression, prices have advanced -tenallv an^^ai. now
in some cases 75 to 100 per cent above low pomt. However,
there aremd cat ons that the country has ample capacity o care
lor Us needs m the immediate and even more remote future.
'^ Iron Age," June 29, 1905-
I76 The Iron and Steel Magazine
Dry Air in the Blast Furnace. — It is worth notice that no
ironmaster at home or abroad has hastened to adopt the Gayley
system. Mr. Windsor Richards said that, on Mr. Gayley's
showing, the output of a single furnace might be augmented by
25,000 tons per year, with a saving of 30,000 tons of coke. But
Mr. Richards added that it remained to be proved that such econ-
omies could be realized in this country. The cooling plant is
'expensive, yet it seems that it would pay for itself in one or at
most two years; and this being so, we might naturally expect
to see the system adopted everywhere. The fact that no such
adoption is going on seems to show that a good deal of incredulity
still exists. We are not surprised. The manufacture of pig
iron has long been a strictly scientific pursuit; and before " im-
provements " are accepted as worth having, they must have some
firm basis of explanation to stand on. Ostensibly the cooling
and drying of air for the blast furnace produces unexplained
economies, and men refuse to invest capital on rule-of -thumb
schemes which appear to be more or less flatly opposed to all
that is believed concerning thermodynamics. The iron ore is
deoxidized by the aid of a high temperature in the presence of
carbon. How the process is affected by a small quantity of steam
gas no one seems to be able to tell us with certainty The heat
saved by the removal of a part of this steam gas is too insignifi-
cant to represent anything like a saving of 20 per cent or so of
fuel. Possibly the result is due to a number of causes, all work-
ing together. For ourselves, until something much more con-
clusive has been advanced than anything yet brought forward
we shall continue to hold that the major portion of the saving
is due to an increase in the density of the blast. If this is not
true, then we must unlearn what has been taught concerning
the cost of dissociation, or take it for granted that the presence
of small volumes of free hydrogen in a blast-furnace hearth is
inimical to economy. ^' The Engineer," June 2, 1905.
Electric Smelting of Iron Ore. — The smelting of iron ore
by electricity is to have a practical working test in Canada.
The Dominion government has appropriated $15,000 for the
purposes of the test, and a building and free power will be fur-
nished for a limited period at the Sault Ste. Marie works. The
test will be under the special charge of M. Heroult, the French en-
Metallurgical Notes and Comments 177
ginecr. who has a wide reputation for his operations and experi-
ments in electrical smelting in France. It is understood also
that some experiments will be made with nickel ores from Sud-
burv. Should the experiments result successfully, the Canadian
government will probably spend a larger sum in erecting a plant
at some point in the vicinity of the iron ranges of western Ontario.
*' Engineering and Mining Journal," June 15, 1905.
Success of Open-Hearth Rails. — The Tennessee Coal, Iron
<t R. R. Co. has for some time had sufficient tonnage on its
books to insure operation throughout the year, and some orders
for 1906 were booked by that company several months ago.
The success of its open-hearth rail is being emphasized by the
fact that the company has very recently taken orders for a con-
siderable additional tonnage for 1906 delivery. The Youngstown
mill is now operating on billets and sheet bars and later rails
and billets will be rolled. The consumption of steel will, there-
fore, be greatly increased as well as pig iron. The idle Bessemer
plant at New Castle is scheduled to resume shortly after the
first of next month. " Iron Trade Review%" June 22, 1905.
Distinctions for a Distinguished Metallurgist. — The honorary
degree of LL.D. was recently conferred upon Henry Marion Howe
both by Harvard University and by Lafayette College. President
Eliot's words in conferring the degree were the following:
" Henry Marion Howe, a Boston Latin School boy. Harvard
Bachelor of Arts and Institute of Technology Bachelor of Science,
an author on copper, iron and steel, distinguished for scientific
imagination and a good English style, professor of metallurgy
at Columbia University, consulting metallurgist, honored by
the profession in England, France, Germany, Russia and his
native land."
Metallographic Gems. — Martensite is described as follows,
in one of our technical contemporaries :
" The ingredient ' martensite ' of steel, though detected and
identified by the microscope, has never been isolated. It seems
to be a specialized interlacing of ferrite and cementite."
The same journal also enlightens us concerning the nature
of the eutectic alloy in steel:
" In the case of steel, the assumed eutectic is cementite,
FcgC (containing about 6.7 per cent of carbon)."
lyS The Iron and Steel Magazine
The Technical Press. — I take this opportunity of expressing
our great indebtedness to the technical press for the wonderful
work they do in keeping us supplied with the most recent
information.
We often wonder at the marvels of daily journalism, but to
me they do not seem to compare with the work done by the
technical press, where the matter must be correct and exact.
The daily press, however essential to our well-being, does work
of a more or less ephemeral nature; that of technical papers is
of a lasting description.
In closing my remarks on this subject permit me to tender
to the technical press and its members my own personal in-
debtedness — which must be shared by many thousands of
my fellow-workers ^ for the great services they are continuously
rendering to us. I frankly confess that to the technical press
I largely owe such progress as I have been able to make in ac-
quiring information. Speaking ex cathedra one can take an
opportunity that but rarely comes of expressing sentiments
on this point, which, in my own case, are very deeply felt.
All honor to the British, American and Continental technical
press for the great work they do in helping on the cause of
progress! From R. A. Hadfield's Presidential Address, Iron
and Steel Institute, May, 1905.
Importance of Research and Invention. — In progressive man-
ufacture, the complexity of which increases year by year, there
is, in addition to the many ordinary difficulties met with, that
of how to solve new problems which constantly present them-
selves. This can be done only by research, which should form
an actual part of industrial operations, and demands almost
as much attention as is devoted to the manufacturing side.
As one who has carried out many researches in metallurgy,
and has also benefited by the scientific investigation of others,
not only in this country but elsewhere, it would be an omission
on my part to neglect to make reference to the practical im-
portance of research. It is more than ever necessary not
to rest satisfied with the knowledge of to-day, nor to think
that this will satisfy the needs of to-morrow. Rapid and
great changes are constantly occurring in metallurgy as in
other branches of scientific knowledge.
Metallurgical Notes and CoiiDUcuts 179
Closely allied to research, though natural^ of a somewhat
different order, is invention. The Hon. Charles A. Parsons,
in his presidential address to the Engineering Section of the
British Association at the meeting in Cambridge last vear,
said that he considered the general encouragement of invention
would be one of the greatest steps in the advancement of the
human race that would take place this century. No doubt he
was right. Professor Perry recently pointed out that the
energy in one ton of coal was equal to the work of 40,000 labor-
ers working ten hours a day. Yet to-day our best steam-
engines only utilize one tenth of this energy. What an avenue
here for research and invention ! Prof. R. H. Smith has
also pointed out, in a commentary regarding this question,
that it is upon invention that the progress, almost the civiliza-
tion, of mankind depends. He also calls attention to the
fact that high scientific training is not necessarily productive
of inventors. Too much scholastic training is, in some minds,
apt to take away originality, and to inculcate dogmatism
rather than progress. From R. A. Hadfield's Presidential
Address, Iron and Steel Institute, May, 1905.
REVIEW OF THE IRON AND STEEL MARKET
A material change in the complexion of the iron and steel
market has occurred since last report. A month ago there was
general dullness, pig-iron prices were declining and prices of
some finished steel products were weakening. The rail trade
alone was showing an improvement. Subsequent develop-
ments have all been favorable. A large tonnage of southern
pig iron has been sold and a moderate tonnage of northern iron.
Prices of southern iron have advanced, while prices of northern
iron have hardened. There has been further good buying of
rails, so that all the available mills are operating, with tonnage
booked until almost the close of the year. In the finished steel
lines which were weakest, merchant pipe, sheets, tin plate and
wire products, there has been a little increase in buying, while
stocks have been greatly reduced and reports are quite encour-
aging as to prospects of good business in August. In the lines
which have been steadily active this year, structural shapes and
plates, the pressure has now reached an acute stage, the large
mills being unable to offer any deliveries at all on new business
or even on new specifications against old contracts, within from
three to six months, while stocks have been so depleted that
some smaller mills, which can make earlier shipments, are easily
securing premiums over the regular prices.
It is quite unusual for the iron trade to show improvement
during July, and the augury for fall business is extremely favor-
able. General business conditions are very good, crop prospects
are excellent, and there is hardly a cloud on the horizon. The
iron trade looks confidently for good business during the second
half of this year, and for better business next year than this.
Pig Iron. — Sales of southern iron during July aggregated
probably over 100,000 tons, or more than the current make.
There was very little buying until the open market had receded
to $11.25, Birmingham, for N°- ^ foundry, whereupon, by mak-
ing inside prices the southern furnaces tempted a large business,
the largest buyer being the cast-iron pipe consolidation, which
I Ho
Rci'icw of the Iron and Steel Market
i»i
took 25,000 tons or more. The prices at which this business
went are not exactly ascertainable, but are supposed to have been
from S10.50 to $11.00. The market then advanced sharply,
most large producers advancing to $11.50 and in many cases
asking $12.00 for fourth quarter, leaving but little iron to be
had at $11.25 ^^*^ nothing at lower figures. Of northern iron
there have been only moderate sales, but the situation is work-
ing out in the direction of large buyers coming into the market
shortly. The lowest prices now named cover only a limited
tonnage, for early shipment, and as soon as this iron is taken the
higher prices already named by some furnaces are likely to pre-
vail. Sales of No. 2 foundry have been made at $14.00, f. o. b.
central western furnace, and possibly lower. On malleable
Bessemer, $14.00 at furnace has been shaded materially, and by
a small amount on standard Bessemer. We quote prices as
follows, the lower figures, being, as noted, on comparatively small
offerings which are likely to be absorbed shortly: F. o. b. valley
furnace: Bessemer, $14.25 to $14.50; basic, $14.00 to $14.25;
No. 2 foundr}^ $14.00 to $14.25; gray forge, $13.50 to $13.75.
Delivered Pittsburg: Bessemer, $15.10 to $15.35; basic, $14.85
to $15.10; No. 2 foundry, $14.85 to $15.30; gray forge, $14.35
to $14.60. F. o. b. Birmingham: No. 2 foundry, $11.25 "to $12.00;
gray forge, $10.25 to $11.00. Delivered Philadelphia; No. 2 X
foundry, $16.25 "to $16.50; standard gray forge, $14.50 to $14.75.
Delivered Chicago: northern No. 2 foundry, $16.25 "to $16.50;
malleable Bessemer, $16.25 to $16.50. Freight: Birmingham
to Pittsburg, $4.35; to Cincinnati, $2.75; to Chicago, $3.65.
Steel. — Buyers' ideas have been that they should be able
to buy steel for third quarter at prices considerably lower than
those which prevailed for deliveries in second quarter, but have
found that very little steel could be picked up at such figures,
the leading interest holding to higher prices and appearing to
have the market well in hand. We would, therefore, quote
Bessemer 4x4 and larger billets at $23.00; sheet bars, long
lengths, $24.50; sheet bars, cut, $25.00, — all f. o. b. Pittsburg,
for third quarter delivery.
Shapes. — There is still more pressure on the structural mills,
the leading interests having specifications which will keep them
busy for from four to six months. As they are behind on these,
they cannot take new business for earlier shipment, and such
1 82 The Iron and Steel Magazine
business is going to some of the smaller eastern mills, generally
at premiums. Regular mill prices were reaffirmed on July 19
at 1.60 cents for beams and channels, 3 to 15 inch inclusive,
angles 2x3 to 6x6 inclusive, and zees.
Plates. — Conditions are much the same as in shapes. There
have been good inquiries for steel cars lately, and another buy-
ing movement may be inaugurated. Prices were reaffirmed on
Julv 19 at 1.50 cents for plates 14 inches wide and under, and
1.60 cents for plates over 14 and not over 100 inches wide, for
tank quality, quarter-inch and heavier, with the usual advances
for other grades and sizes.
Merchant Bars. — Specifications on steel bars have improved,
there is less shading by overloaded jobbers, and some small
business is being placed with the mills at full prices, while some
large consumers have begun considering the question of material
for next year. In general, the market can be quoted at the regu-
lar price of 1.50 cents, half extras, f. o. b. Pittsburg, carload
and larger lots. The iron-bar market is quiet and we continue
to quote 1.55 cents to 1.60, Pittsburg. The Chicago market,
after taking a dip to 1.45 cents, has returned to our last quota-
tion of 1.50 cents to 1.55, there being difficulty in doing the
lower figure.
Sheets. — Wage matters with the Amalgamated Association,
which controls about one fourth of the leading interest's sheet
mills and over half of the independents, have been satisfactorily
adjusted for the year beginning July i, on the basis of former
rates, with the output limit entirely abolished. Most of the
union mills are still closed, and some of the non-union mills.
Demand continues fairly good and stocks are becoming ex-
hausted, so that mill business should improve during August.
We quote 2.35 to 2.40 cents on black and 3.40 to 3.45 cents on
galvanized, No. 28 gauge, for ordinary carloads. On a desirable
order about 5 cents a hundred less can be done.
Scrap. — The decline in scrap appears to be over, the market
having shown a slight hardening tendency. Small lots of heavy
melting stock can be picked up occasionally at $14.00, delivered
Pittsburg, but any fair-sized quantity would bring $14.25 to
$14.50. Other prices are approximately as follows: Sheet scrap,
$12.50 to $13.00; old car wheels, $14.50 to $15.00; cast borings,
$8.00 to $8.25 — all gross tons, delivered Pittsburg.
STATISTICS
steel Production in Canada.* — The following table gives
the production of all kinds of steel ingots and castings in Canada
from 1894 to 1904, in gross tons:
1
Years Gross Tons
1
Years
Gross Tons
Years
Gross Tons
1S94 . . .
1895 . . .
1896 . . .
1897 • . •
25,685
1 7,000
16,000
1 8,400
1898 . . .
1899 . . .
1900 . . .
1 90 1 . . .
21,540
22,000
23.577
26,084
1902 . . .
1903 . . •
1904 . . .
182,037
181,514
148,784
The following table gives the production of all kinds of
iron and steel rolled into finished forms in Canada from 1895
to 1904:
Years
Gross Tons
Years
Gross Tons
Years
Gross Tons
1895
1896
1897
1898
66,402
75.043
77,021
90,303
1899
1900
I9OI
1902
110,642
100,690
112,007
161,485
1903
1904
129,516
180,038
The production of Bessemer and open-hearth steel rails in
1904 amounted to 36,216 gross tons, against 1,243 tons in 1903 ;
structural shapes, 447 tons, against 1,983 tons in 1903; cut nails
made by rolling mills and steel works having cut-nail factories
connected with their plants, 99,000 kegs of 100 pounds, against
118,686 kegs in 1903; plates and sheets, 3,102 tons, against
2,450 tons in 1903; all other finished rolled products, excluding
muck and scrap bars, blooms, billets, sheet bars and other un-
finished forms, 135,243 tons, against 118,541 tons in 1903. The
total quantity of all kinds of steel and iron rolled into finished
forms in Canada in 1904 amounted to 180,038 tons, against
* " The Bulletin," American Iron and Steel Association, June i, 1905.
183
184 The Iron and Steel Magazine
129,516 tons in 1903. Of the 180,038 tons of finished iron and
steel reported for 1904, about 126,850 tons were steel and 53,188
tons were iron.
On December 31, 1904, there were 18 completed rolling mills
and steel works in Canada. In addition, 3 plants were being
built, and 2 plants were projected. Of the completed plants, 2
were equipped for the manufacture of steel castings only, 5 for
the manufacture of Bessemer or open-hearth steel ingots and
rolled products, and 11 for the manufacture of rolled products
only. Of the building plants, one was being equipped for
the manufacture of steel castings by a special process, one for the
manufacture of open-hearth steel ingots only, and one for the
manufacture of merchant bar iron, railway spikes, etc. One of
the projected plants is to be equipped for the manufacture of
skelp and bar iron and the other for the manufacture of wire
rods.
Of the 18 completed rolling mills and steel works in Canada
on December 31, 1904, 3 were located in Nova Scotia, 5 in
Quebec, 9 in Ontario, and i in New Brunswick. The building
plants are in Nova Scotia, Ontario and Manitoba, and the
projected plants are in Ontario.
We are ofhcially advised that the production of iron ore
in Canada in 1904 amounted to 312,286 gross tons, against
235,977 tons in 1903, and that the production of coal in Canada
in 1904 amounted to 6,705,232 gross tons, against 6,824,999
tons in 1903. The figures for 1904 are subject to revision, but
are substantially correct.
Iron and Steel Imports and Exports.* — During the first
ten months of the present fiscal year, which ends on June 30, this
country imported iron and steel and manufactures of iron and
steel of the value of $18,488,000, against $23,077,552 in the
corresponding months of the preceding fiscal year, and it ex-
ported iron and steel and manufactures thereof of the value of
$110,788,570, against $89,107,854 in the corresponding months
of the preceding fiscal year. Of the imports in the last ten
months, ending with April, we note 81,028 tons of pig iron and
3,496 tons of iron and steel rails, against 181,889 and 37,632
tons respectively in the ten months of the fiscal year 1904, and
* "The Bulletin," American Iron and vSteel Association, June i, 1905.
Statistics 185
of the exports in the last ten months we note 42,534 tons of pig
iron and 372,380 tons of steel rails, against 29,945 and 68,405
tons, respectively, in the ten months of the fiscal year 1904.
German Steel Production.* — The corrected report of the
German Iron and Steel Union gives the production of steel in
Germanv for the full year as below, in metric tons:
Acid Basic Total
Converter ingots 423,74^ 5»525'429 5'949.i7i
Open-hearth ingots 130,546 2,697,760 2,828,306
Direct castings 56,409 96,405 152,814
Total 610,697 8,319,594 8,930,291
Total, 1903 613,399 8,188,116 8,801,515
The total production last year showed an increase of 1 28,776,
or 1.5 per cent, over the previous year. It will be noted that
93.2 per cent of the total was basic steel, and that 55.4 per cent
was made in the converter.
* " Engineering and Mining Journal," June 22, 1905.
RECENT PUBLICATIONS
Statistics of the American and Foreign Iron Trades for 1904.
136 6 X 9-in. pages; paper covers. The American Iron and
Steel Association. Philadelphia. 1905. Price, $5.00. — This
is Mr. Swank's thirty-third annual report, and in degree of ex-
haustiveness and general excellence it even excels the high
standard of his previous reports. It was presented to the mem-
bers June 10, and the collection of such a vast amount of in-
formation at so early a date is in itself a feat deserving of high
commendation. The American Iron and Steel Industry is cer-
tainly fortunate to have had for so many years the services of so
eminent and capable a statistician.
" This report embraces all the leading features of previous
reports and also many new features. The statistics of produc-
tion of iron and steel are full and complete. Tables are given
which show our annual imports and exports of iron and steel,
tinplates, iron ore, etc. Details are given of the shipments of
iron ore from the Lake Superior and other mines, the imports
of Cuban iron ore, the prices of Lake Superior iron ore, the ship-
ments and prices of Connellsville coke, the imports and exports
of coal and coke, the tonnage of steel vessels built in 1903 and
1904, immigration in 1904 and previous years, etc. Our price
tables must especially commend themselves to all who are con-
nected with the iron trade. So also must the tables relating to
our production of steel, which give in detail the annual growth
in recent years of every kind of steel, including all kinds of steel
castings. Our statistics of rail production in 1904 are given in
more than usual detail. There are other new features. Tables
are given which show the production of leading iron and steel
products, iron ore, etc., by the United States Steel Corporation
and by independent companies in 1902, 1903 and 1904. Cana-
dian iron and steel statistics are full and complete. Detailed
statistics of the iron and steel industries of Great Britain, Ger-
many, France and Belgium in 1903 and 1904 are also given.
186
Recent Piihlicatioiis 187
The report closes with statistical tables of the world's production
of iron and steel and iron ore and coal. The Necrological Record
is continued.
'* Following the report proper for 1904, there will be found a
statistical abstract of all trustworthy statistics, mainly of our own
collection, relating to every branch of the iron trade and going
as far back in each instance as such statistics are available. The
large number of tables we give in this abstract and their compre-
hensive character combine to make this feature of the present
report a most valuable contribution to the history of the Ameri-
can iron trade and to national and international economic liter-
ature. Tables showing the prices of Bessemer rails in this
country and in Great Britain for a long series of years and also
showing the miles of railroad in operation in the United States
since 1830 and the displacement of iron rails by steel rails since
1880, will be found valuable for reference by railroad men."
Constructional Steel Work, hy A.^ .Ya.rns\^orth.. 248 6X9-
m. pages; over 100 illustrations. Charles Griffin & Co. London.
1905. Price, 105. 6d. — This book consists of ''Notes on the
Practical Aspect and the Principles of Design, together with an
Account of the Present Methods and Tools of Manufacture."
The author takes the stand that '' there are in this country
[England] thousands of civil and mechanical engineers, archi-
tects and surveyors who find they have, in increasingly numerous
cases, to design and superintend erections involving in some
form or other the employment of mild steel work, and that it
has become imperative that every designer should know some-
thing about it. The present work is an effort to afford to de-
signers generally an indication of what they should seek to
embody in their creation ; it is hoped that it may be also a hand-
book for those practically engaged in the trade." Part I deals
with practical designing and is subdivided into eleven chapters,
while Part II is devoted to practical shop work and contains
nine chapters.
When referring to purely metallurgical questions, the author
is evidently not upon his own ground. On page 17 he divides
the processes employed in England for the manufacture of con-
structional steel into '^ (i) the basic, (2) the Siemens open-
hearth and (3) the Siemens-Martin open-hearth acid processes."
1 88 The Iron and Steel Magazine
It must be presumed that he means (i) basic open-hearth pro-
cess, (2) acid (?) open-hearth processes using pig and ore, and (3)
acid open-hearth process using pig and scrap. A Httle later,
however, he refers to the second process as " open-hearth basic,"
adding that " through recent discoveries, it has not yet taken unto
itself a definite name." What the author means by this will
not, we think, be readily understood. On page 24 he gives the
following figures as representing the allowable limits of foreign
elements in mild steels " sanctioned by the modern metallur-
gical practice": Carbon .02 to .06 per cent, manganese .40
to .90 per cent, silica (meaning silicon) .00 to .05 per cent, sulphur
.00 to .04 per cent and phosphorus .02 to .10 per cent. It w^ould
be interesting to know how much mild steel is being manufac-
tured containing .02 per cent carbon, or indeed .06 per cent, and
if this represents the upper limit, few mills, we think, would care
to furnish it. The lower limit for silica (!) is surely low enough
as well as the upper limit, and the same may be said of sulphur.
In the case of phosphorus and manganese the author gives more
rational figures.
Steam and Steam Engines, including Turbines and Boilers, by
Andrew Jamieson. Fourteenth edition. 780 4^ X 72-i^-
pages; numerous illustrations. Charles Griffin & Co. Lon-
don. 1904. Price, 105. 6d. The fact that this book has
reached its fourteenth edition is conclusive evidence of its value
and popularity. In this recent edition the author has added
two lectures dealing with steam turbines and bringing that im-
portant subject up to date. Pyrometry and calorimetry are
also treated at greater length than in previous editions, while
increased space is devoted to thermodynamics and to wet, dry
and superheated steam. The subject is treated in the
form of lectures and each lecture is followed by a series of ques-
tions, there being 886 questions in the book. While it is pri-
marily written for students, it is also a reference book of great
value.
Gas Producers, by W. A. Toakey. 137 4.^ X 6-in. pages;
illustrated. Percival Marshall & Co. London. 1905. Price,
15. — The increased use of the gas engine has brought gas
producers to the front as never before. New producers are
Recent Publications 189
almost dailv described in the technical press and many articles
have appeared lately descriptive of producer gas and its genera-
tion. The little book we have before us is, therefore, a timely
one. It should prove of service to many because of its brief
and clear treatment of the subject. The following titles of the
chapters will show the ground covered : Producer Gas: Method
of Generation, Composition and Qualities; The Cost of Producer
Gas; Gas from Bituminous Coal: the Mond Principle; Gas
from Non-Bituminous Coal: the Dawson Principle; Gas from
Waste Wood, etc.: the Riche Principle; Gas from Iron Bitu-
minous Coal: the Suction Principle; Hints for Erectors and
Attendants ; General Remarks ; Suction Gas Plants : illustrated
description of the designs offered by the leading makers.
Testing of Electro-Magnetic Machinery and Other Apparatus.
Volume I. Direct Ctirrents, by Bernard Victor Swenson and
Budd Frankenfield. 420 4^ X 8|-in. pages; illustrated. The
Macmillan Company. New York. 1904. Price, $3.50. — The
authors in their preface state that this' book is intended for use
as a college textbook and also as a work of reference for the
engineer. The field covered by the present volume is that of
direct current electromagnetic machinery and apparatus, and
is almost exclusively confined to dynamo-electric machiner}-.
The second volume (which is in course of preparation) will deal
with alternating current machinery and apparatus. In the
present volume ninety -six experiments are described, each one
being self-contained. The authors have undoubtedly bestowed
a large amount of work upon this book and the indications are
that the subject has been treated with great care and much
authority.
Fire Assaying Notes, by F. P. Dunnington, School of Analy-
tical Chemistry, University of Virginia. 18 6 X 9-in. pages.
Eschenbach Printing Company. Easton, Pa., 1905. — This
pamphlet consists of some lecture notes in which the elements of
fire assaying are briefly presented.
Automobiles : Vapeur, Petrole, Electricite, by H. Rodier.
175 9 X i2;\-in. pages; illustrated. Paris. 1905. Price, $4.25.
— The automobile being essentially a machine of French origin
190 The Iron and Steel Magazine
and having attained its highest perfection in France, it is fitting
that the first exhaustive treatise on that subject should be written
by a French expert. This book should be of great value to the
constructor of automobiles. It is well printed on excellent paper
and well illustrated.
German Technical Words and Phrases, by C. A. Thim and
W. Von Knoblauch. 214 3 X 5^-in pages. F. Marlborough
& Co. London. 1904. Price, cloth, 25. 6(i. ; leather, t^s.
6d. — This little book is an English-German and German-English
dictionary of technical and business terms and phrases used in
commerce, arts, sciences, professions and trades.
i
PATENTS
UNITED STATES
791,170. IMaxufacturk of Steel. — James Vernon, Newton
Stewart, Scotland.
791,189. Manufacture of Manganese-Steel Rails or Shapes.
— Robert A. Hadfield, Sheffield, England.
791,193. Machine for Straightening Tubes, Shafts, Bars or
THE Like. — Otto Heer, Dusseldorf, Germany.
791,461. Gas-Producer Apparatus. — Carleton Ellis, New York,
N. Y.
791,494. Apparatus for Magnetic Separation. — Clarence Q.
Payne, Stamford, Conn., assignor to the International Separator Com-
pany.
791,833- Roller-Mill. — Simon Snyder, Muncy, Pa., assignor
to Sprout, Waldron & Co., Muncy, Pa.
791,928. Process of Treating Ferruginous Ore for the Manu-
facture OF Iron and Steel Therefrom. — Montague Moore, Melbourne,
and Thomas J. Heskett, Brunswick, Victoria, Australia.
791,940. Feeding Device for Slabs, Billets, Etc. — Casimir
von Philip, Bethlehem, Pa., assignor to Bethlehem Steel Company, South
Bethlehem, Pa.
791,958. Automatic Blast-Temperature-Regulating Valve. —
Samuel W. Vaughen, Lorain, Ohio, and John W. Cabot, Johnstown, Pa.
792,047. Apparatus for Changing Blast-Furnace Bells. —
Walter Kennedy, Allegheny, Pa.
792,179. Tuyere Iron. — Charles O. Swenson, Minneapolis, Minn.
792,231. Art of Cross-Rolling Tubular Bodies or Blanks in
A Heated State. — John H. Nicholson, Pittsburg, Pa., assignor to
National Tube Company, New York, N. Y.
792,440. Apparatus for Treating Ferruginous Ore for the
^LA.NUFACTURE OF Iron AND Steel Therefrom. — Montaguc Moore,
Melbourne, and Thomas J. Heskett, Brunswick, Victoria, Australia.
792,591. Metal-Working Machine. — James Hartness, Spring-
field, Vt.
792,619. Portable Furnace for Melting Steel or Other
Metals. — Louis Rousseau, Argenteuil, France.
792,630. Ingot vStripper. — Clarence L. Taylor, Alliance, Ohio,
assignor to the Morgan Engineering Company, Alliance, Ohio.
792,642. Melting Furnace. — William E. Williams, Chicago, 111.
191
192 The Iron and Steel Magazine
792,681. Furnace-Charging Box. — Clarence L. Taylor, Alliance^
Ohio, assignor to the Morgan Engineering Company, Alliance, Ohio.
792,735. Furnace-Filling Apparatus. — John W. Seaver, Cleve-
land, Ohio, assignor to the Wellman-Seaver-Morgan Company, Cleveland,.
Ohio.
792,914. Method of Making Open-Hearth Steel. — Niven
McConnell, Pittsburg, Pa.
792,992. Mold for Producing Tuyere-Molds. — Samuel A..
Kelly, Pittsburg, Pa.
793,027. Indicating Attachment for Rolling-Mills. — Harry
H. Burton, Southford, and James L. Burton, New Britain, Conn.; said
James L. Gurton assignor to said Harry H. Burton.
GREAT BRITAIN
12,785 of 1904. Gas Smelting of Iron Ores. — E. Fleischer,.
Dresden, Germany. Smelting iron ores in inclined revolving tubular
furnaces by means of water gas.
16,412 of 1904. Case-Hardening Steel. — Cyanide Gesellschaft,.
Berlin, Germany. Case-hardening iron and steel by packing with organic
refuse containing a large proportion of nitrogen and heating to a high
temperature.
9,202 of 1904. Smelting Iron. — R. M. Daelen, Dtisseldorf, Ger-
many. Smelting fine ores by briquetting mixtures of the ores and finely
divided carbon, and after reduction charging them into a molten bath of
iron to prevent reoxidization of the spongy iron formed.
27,839 of 1904. Utilizing Cupola Gases. — A. Bayot, Marly,.
France. Arrangement of pipes so as to catch the gases such as carbon
monoxide and carbohydrates given off from foundry cupolas and utilizing
them for heating the cupolas.
9,836 of 1904. Coating Iron. — A. Levy, Paris, France. Forming
permanent protective coverings for iron by first coating with zinc and
afterward with tin, lead or copper.
12,817 of 1904. Tempering Steel. — S. N. Brayshaw, Manchester.
Tempering high-speed tool steel, by first heating in furnace, then quench-
ing in a bath kept at about 800° C. and afterward quenching in water.
16,214 of 1904. Removing Phosphorus from Iron Ore. — W.
Simpkin and J. B. Ballantine, London. Removing phosphorus from
finely ground iron ores by leaching out with dilute acid solutions.
16,276 of 1904. Steel Making. — W. Kaufman, Vienna, and A.
Bauvier, Grenoble, France. Improved method of introducing carbide
of silicon into steel, so that it shall not be decomposed.
R. W. RAYMOND
SEE PAGE 250
The Iron and Steel Magazine
" J'e veux au mond publicr
d'une plume de fer sur un papier d'acicr."
Vol. X September, 1905 No. 3
DESCRIPTIVE METALLURGY OF IRON AND STEEL *
By SAMUEL GROVES
IXTRODUCTORY
TX the primitive ages, when our ancestors dwelt in caves, they
hunted the buffalo and the bear with bow and arrow and
spear, tipped with sharpened flints and stones. It was this
unique faculty of being able to invent and form tools and weapons
of defense that " gave man dominion over the fowls of the air,
the fish of the sea and everything that creepeth on the face of
the earth." Moreover, this capacity to give new shape and
form to existing materials makes an absolute line of demarka-
tion between man and the lower animals. Monkeys will throw
down cocoanuts from the trees, then pick up a stone and crack
the shells; but they always take something ready to their hand
— never make a hammer, hatchet or knife. Elephants tear
down branches from the forest trees to drive away the torment-
ing flies, or to protect them from the burning rays of the tropical
sun; but who ever heard of an elephant making an umbrella?
There is not an authenticated instance in the history of the
world of an animal lower than man inventing and making a
tool or implement to aid his natural powers. By instinct — "a
propensity existing prior to experience and independent of in-
struction"—the beaver builds his dam, the bee his honevcomb.
* " The Canadian Engineer," July, 1905.
This article is copyrighted in the United States and is reproduced
here through the special permission of Mr. Samuel Groves, editor of " The
Canadian Engineer."
194
The Iron and Steel Magazine
and the robin his nest, just as they did at the earliest dawn of
civilization. The cave, tent, cabin, cottage, house and palace
indicate the progress made by the human race. By man's
unique power of invention the desert has been made to blossom
as the rose.
The earliest records of prehistoric times show Palaeolithic
man dwelling in hillside caverns, supplementing his natural
powers by making stone arrowheads and spear points to protect
Fig. I. A Family of the Stone Age
him from the great carnivora, or to kill wild deer for food, and
shaping stone hatchets, knives and ham.mers for cutting up the
flesh and breaking the bones. Hence this first period of man's
existence has been called the
St ONE Agk
Then came the discovery of copper, which was hammered
into ornaments and x^arious articles of domestic use, but was too
Descriptive Metallurgy of Iron and Steel
195
soft for fine-edged tools and weapons of defense. It was found,
however, that by mixing the ores of oxidized copper and tin
together, and melting them over a hot charcoal fire, a new
metallic allov resulted, which could be poured as a liquid into
molds, making castings of all shapes and sizes, having any degree
of hardness; in fact, a mixture of two parts copper and one of
tin makes an alloy so hard that it cannot be cut by ordinary tool
steel. It was further discovered that if the bronze was made
red hot, then siiddenly plunged into cold water, it becomes com-
paratively soft and ductile, and could be hammered into any
shape, but if made red hot again, and allowed to cool slowly,
Fig. 2. A Founder's Workshop during the Bronze Period
would once more regain its original hai"dness. The native workers
of bronze in India still take advantage of this diverse process,
for it is the very way in which they make their cymbals and tom-
toms to-day.
We thus see that in bronze, Neolithic man discovered a
metallic composition admirably adapted to his nascent indus-
trial skill, for without much trouble he could pour it as a
liquid into molds having the form of hatchets, poinards, swords,
agricultural implements and mechanical appliances of all kinds.
In this way began the art of founding in metals. At what
precise date and in which part of the globe this simple but im-
196 The Iron and Steel Magazine
portant invention was made is enveloped in mystery. The
earliest written record is found in Genesis 4:22, where Tubal-cain
is described as " an instructor of every artificer in brass and iron,''
while tradition declares that bronze was first introduced into
Central Europe by Phoenician traders about 1500 B.C.
This e[)0ch, or second period in the story of mankind, is
known as the
Bronze Age
But although bronze was a great advance over stone as a
means of aiding the natural powers of primeval man, yet, as
organized communities took the place of the patriarchal family
and tribe, and the wants of the social organism became neces-
sarily more varied and complex, it was found that copper and
tin, and the bronze alloy made therefrom, were not elastic and
hard enough, and the supply of their ores not plentiful enough,
to meet the increasing demands of the newly formed societies.
And since " Necessity is the mother of invention," and the art
of metallurgy had made great progress during this epoch, as the
unearthed prehistoric relics of the period show, it is not surpris-
ing that Tubal-cain, or some other skillful founder in " brass,"
tried the experiment of smelting in a hot charcoal furnace the
red oxides of iron w^hich they found in abundance cropping out
of the hillsides, and succeeded in easily producing lumps of mietal,
unalloyed, which could be hammered into knives, hatchets,
scythes, plowshares, etc., for domestic use or cultivation of
their lands in the " piping times of peace," and swords, spears,
armor, etc., for use in the chase, or stirring times of war and
conquest.
With the advent of iron began a new era in the social evo-
lution of mankind, known as the
Iron Age
As in the case of bronze, so in the case of iron, the exact
time and place of the discovery is unknown; but archaeological
remains, gold trinkets, bronze coins, ivory ornaments, iron
swords, scythes, etc., together with rudely built smelting fur-
naces, found in the hillsides and prehistoric tombs of Hallstadt,
Austria, and Jura Mountains of Switzerland, clearly prove that
2000 B.C. a Gallic race inhabited these regions of Central Europe,
Descriptive Metallurgy of Iron and Steel
197
who, by taking advantai^e of the discovery of iron, and trading
their manufactures with the Pha'nicians, enjoyed material com-
forts and luxuries and developed skill and taste in the cultiva-
tion of the arts of life, far in advance of the rugged nomadic
tribes and people bv whom they were surrounded, bearing out
the famous dictum of Thenard, the chemist, that we may judge
of the state of civilization of any nation by the degree of per-
fection at which it has arrived in the workmanship of iron.
The method by which these men of the early Iron Age
extracted metallic iron from mineral ore is graphically illus-
trated in Fig. 3. This picture was drawn from a model in relief.
Fig. 3. Primitive Furnace for vSmelting Iron
prepared in 1S66 by a learned Swiss engineer, M. Quiquerez,
and designed from many specimens found in Hallstadt, Austria,
and in the Bernese Jura. The furnace consisted of a cavity in
the hillside, covered in with a concave wall about nine feet high,
plastered with fireclay, and surmounted with a conical chimney.
Steps made of rough stone were arranged on each side of the
mound to enable the workman to climb on top in order to charge
with ore and fuel. On the right-hand side is a heap of charcoal
for fuel, while on the left is a store of ironstone, enclosed in a
pen formed of long, wooden logs. In the foreground is a heap
of scoria, hammer slag and scale, dropped as debris in the pro-
19 S The Iron and Steel Magazine
cess of hamnieririg the crude metal. A workman is pulling a
cake of iron out of the ashes; another is hammiering on the anvil
a piece of spongy iron, just taken from the furnace. In all these
researches no trace was found of the use of bellows, natural
draught only being used at this period; hence there is no proof
that joiinding in iron existed in prehistoric times. To fuse iron
ore and reduce it to a liquid condition requires a temperature
of some 2200° F., and this high temperature is not attainable
in the natural draught furnace described. In these ancient
furnaces the iron in the mineral ore was only reduced to a red-
hot spongy state, and dropped down among the ashes in the
form of pasty lumps of malleable iron, weighing from twelve
to sixteen pounds, which were worked on the anvil Vjy artisans
skilled in the craft of Tubal-cain.
So far we have followed the footprints of primitive man
up through the mists of antiquity, guided only by the evidence
furnished in roughly chipped Hints and stones, cunningly worked
bronze tools and ornaments, rudel}^ formed iron weapons, im-
plements, etc., found in deep caverns, embedded in solidified
m.ud or under alluvial deposits of past geological ages.
It is not until 1050 B.C. that we cross the threshold of
history, and enter the domain of scientific fact. The first historic
record is found in Solomon's famous miessage to Hiram, king of
Tyre (2 Chron. 2:7), where he requests the Phoenician monarch
to assist him in building the Temple of Jerusalem, in the golden
age of Judea. Said he:
" Send me now therefore a man cunning to work in gold,
and in silver, and in brass, and iron."
The next instance is found in Homer's " Iliad," written
about 850 B.C.:
" And the Greeks bought wine for shining steel, and some
for sounding brass." — {Book VII.)
iVnd where the Trojan captive spy, Dolon, tries to bribe
the Greeks:
"... He, weeping, offered a wealthy ransom for his life,
and told them he had brass, much gold, and iron, that fit for
many labors was, from which rich heaps his father would a
wondrous portion give." — {Book X.)
From this time forward the material progress and civili-
zation of the human race, especially in Europe, was greatly
Tlii' Mdir.ijjctiirc and Characteristics of Wrought Iron 199
accelerated, and largely through the use of iron, which has feli-
citously been called the " king of metals."
{To be continued)
While it is the author's intention to incorporate into this
series everything of importance connected with the metallurgy
of iron and steel, his chief aim will be to make the description
as graphic and lucid as possible. In the matter of illustrations
neither time nor money will be spared to make these of super-
lative interest. — Editor.
THE MANUFACTURE AND CHARACTERISTICS OF
WROUGHT IRON *
By JAMES P. ROE, Pottstown, Pa.
I. IXTRODUCTIOX
npHOSE who deem the subject of this paper an old and super-
-'' seded one may recall with advantage the words of the great
proverb-maker, bidding us to seek the new in the ashes of the
old.
The manufacture of wrought-iron began with the small
hearth built upon an eminence, and relying on the wind for
blast. Next came, either the hearth about 12 inches in diameter
by 2 feet in height, blown by a man operating two goatskins
for bellows, or a hearth of greatly varying diameter and about
10 feet high, depending for its blast upon natural draught. The
former appeared in India, the latter in Burmah; and both were
in operation until quite recently, though their origin is lost in
prehistoric times.
The Catalan forge was the evolutionary descendant of the
foregoing, and the parent of the Blase-ofen, which in turn led
to the high bloomery or " Stiickofen," the father of the blast
furnace, as a producer of cast iron.
The production of cast iron checked the growth of the direct
processes and led to the introduction of various hearths, such
as the Walloon, for low-silicon irons, and the South Wales
* American Institute of Mining Engineers, May, 1905, meeting.
200 The Iron and Steel Magazine
process, combining the run-out fire with the finery, for higher-
siHcon irons. Some of these are still in operation in locaHties
possessing the necessary raw materials, and their product is
used for purposes requiring wrought iron of exceptional quality,
the supply of which is still smaller than the demand.
We have to look back but a century and a quarter to see
the introduction of the puddling process by Henry Cort ; a pro-
cess which, reinforced by the iron-oxide bottoms and fettling
of Hall, was not less revolutionary in character, and more revo-
lutionar}^ in effect, than any other single step in the metallurgy
of iron.
Cort w^as the first to avoid contamination of the product
through contact with the fuel, thus increasing the available
fuel supply, and permitting a limitless variety of forms for the
furnace. In conjunction with Hall's improvements, his pro-
cess was the first, and still remains the only one, w^hich can fur-
nish a good product from pig irons relatively higli in silicon,
phOvSphorus and sulphur. Other processes can only take care
of one, or at the most two, of the mictalloids named. The
vastly increased supplv of iron ore thus made available had
much to do with the sujjremacy in the iron manufacture so
long held bv Great Britain, while it promoted also the indus-
trial progress of other nations.
The circumstance that pig irons high in phosphorus and std-
phur may be used in puddling has m.ore than an immediate
commercial significance. In view of the immense and ever-in-
creasing consumption of iron, there must be a limit to the purer
ores and fuels available; and a process in which not only phos-
phorus and sulphur can be eliminated to any desired extent,
but the finished product of which may, if wished, retain four
times as much phosphorus as good steel, and yet be a thoroughly
reliable material for use, is certain to play a part in the fate of
nations.
The puddling process as carried out by Cort, and later
known as " dry puddling," was effected upon a sand bottom and
was open to the objections: (i) that the oxidation of the mietal-
loids was effected by atmospheric action, resulting in high loss;
(2) that the volume of cinder as a receptacle for oxidized impur-
ities was small, and hence the range in character of pig iron
available was limited; and (3) that, as a consequence, various
TJie Mafuifaciiirc an J Characteristics of Wrought Iron 201
Icinds of so-called " physic " were employed, resulting in an un-
certain product.
The puddling process of to-day, known as the " pig-boiling
process," and invented about 1830 by Joseph Hall, consists in
the use of relatively infusible oxides of iron for the bottom- and
side-fettling of the hearth. Its results are smaller loss of metal,
greater output of more certain character, the possibility of using
gray pig, and the furnishing of the " physic " that had so long
been sought.
The process is both simple and beautiful. Simple, in that
the fettling or "fix " forms a reservoir of relatively pure oxide
of iron which is automatically drawn upon by the cinder of the
bath for correction: that is, the higher the silica of the bath
the greater its draught upon the relatively basic sides and bottom,
to produce a final cinder, low and constant in silica, which is
the factor controlling the phosphorus and sulphur of the prod-
uct from a given grade of pig iron; beautiful, because the
reactions, or their immediate effects, and the crystallization, are
all plainly ^'isible and well defined.
The process consists, essentially, in the removal from molten
pig iron of nearly all its carbon and silicon, and most of its
phosphorus and sulphur, by agitation in the presence of a suit-
able cinder, and gases of the right composition and tempera-
ture; and, finally, by crystallization, due to the greater infusi-
bility of the iron as the metalloids are oxidized.
It demands of the puddler a reasonable degree of skill, and
a kind of labor which, though healthy, is very severe. The
necessity of such labor has long been recognized as the great
practical difficulty and expense of the process; and numberless
mechanical devices have been proposed for its diminution, but,
thus far, without any marked degree of success.
II. Mechanical Pi^ddlixg
The attempts at mechanical puddling may be summarized in
four general classes :
I. The use of hollow rabbles operated by hand, through
which air or steam (sometimes both) were forced into the bath.
These involved complications, a necessarily high loss of iron,
.and little, if any, saving of ]aV)or.
20 2 The Iron and Steel Magazine
2. Stationary furnaces, such as Clough's mechanical pud-
dler, resenibHng in general outline the ordinary puddling fur-
nace, with rabbles operated mechanically. The objections to
this type are its complication, and the fact that it is applicable
only to the puddling period proper, and is in the way during
the operations of fettling, charging, balling and drawing. In
fact, it cannot satisfy the proper requirements for a success-
ful mechanical device, which demands either constant motion,
or a constant C3^cle of motions, which, in turn, call for uniform
shape and conditions; whereas, in a puddling furnace, these
are exev varying.
3. Revolving furnaces, represented by two distinct types,
the Danks and the Pernot, and their respective modifications.
a. The Danks furnace consists of a stationary fire chamber,
a cylindrical hearth revolving about a horizontal axis, and a
removable flue section, giving access to the hearth for charging
and drawing. The advantages of this furnace, which came so
near success as to mislead many acute experts, consisted in re-
ducing the labor of puddling and balling, thoroughly agitating
the heat, and working on a hot bottom. Its disadvantages
were: the excessive time and labor expended in fettling, the
difficulty of drawing the heat, the troubles connected with the
•lining of the revolving portion alternately forming the roof and
the hearth, and, worst of all, the balling of the heat before all
the iron was " ready," which produced blisters. The product
is only available as open-hearth resmelting-stock, for which
purpose it is used, in small quantities, to-day.
b. The Pernot furnace has a hearth like a frying-pan, which
revolves round an axis slightly removed from the vertical. This
slight inclination failed to impart sufficient agitation to the bath,
and collect the crystals into a ball. Hence the rabble and pad-
dle were resorted to, which, in conjunction with the same means
of fettling and drawing, as in the ordinary puddling furnace,
put the Pernot upon the same plane, plus some complications.
4. The oscillating furnace, operating on either a longitudinal
or transverse axis, which held the germ of successful mechani-
cal puddling. Some furnaces of this type, however, were ham-
pered by too narrow mechanical limitations, others were too
complicated, and all of them have hitherto proved incapable of
effecting the whole of the process.
The Maiittfacturc and ( Jiaracteristics of Wrought Iron 203
The advances in puddling thus sketched have been disap-
pointing, and not at all in keeping with the progress achieved
in blast-furnace practice and steel making. This may be ex-
plained in part by the \'ast technical knowledge applied in re-
cent years to the latter lines, leaving the puddling mills largely
to the " practical " men, who have worked their way from
puddlers to foremen, and, possibly, to superintendents, without
becoming emancipated from the traditions to which they were
born.
III. Reactions of the Puddling Process
As already remarked, the puddling process permits the use
of irons of greatly varying composition. My experience covers
irons containing as much as 3 per cent of silicon and of phos-
phorus, 2.5 per cent of manganese, and 0.35 per cent of sul-
phur, though, of course, not all in the same pig iron. In all
these cases, iron was produced that v/ould weld freely, and
showed no cold- or red-short tendency. It is not my intention
to convey the impression that such extreme irons are desirable,
on the contrary, they are costly in time, labor and iron loss.
For rapid work, good yield, and fitness for ordinary uses, a de-
sirable composition would be about i per cent of silicon, a some-
what smaller percentage of phosphorus, o.i per cent of sulphur
and 0.5 per cent of manganese. For special grades, it is neces-
sary to select the pig in each particular case.
The various iron oxides, charged as fettling, flux and oxidiz-
ing agents, should be characterized, in the order named, by in-
fusibility, fusibility and the ability to be readily reduced, for
the part they respectively play in the operation. The fiux, as
the material directly forming the cinder, must he present in
sufficient quantity to receive the oxidized metalloids, and low
enough in silicon to retain them as the temperature of the bath
increases. This cinder is the vehicle for oxygen, whether sup-
plied by roll-scale, ore, or the oxidizing gases of the furnace; it
protects the puddled mass from undue oxidation by the gases;
and, finally, it forms the welding cinder. Its composition is
changed during the process by additions, by reactions upon the
fettling and the bottom, and by the condition of the fire and
position of the damper.
The control of the volume of cinder, in the various stages of
204 The Iron and Steel Magazine
puddling, up to the finished product, may be mentioned here.
Throughout all these stages it is primarily controlled by its
composition as effecting its fusibility. In the ball when drawn
it is controlled largely by temperature at the time of going on
high-boil and at the time of drawing, this temperature, at both
periods, being chiefly governed by the damper. In the shingled
ball it is controlled by the amount of work or pressure, the
cinder being generally smaller in amount when a hammer
than when a squeezer is used. In the puddle-bar, and in the
finished product, it is dependent upon the reduction in area and
the rapidity of this reduction. These conditions put the pro-
cess thoroughly under the control of the operator.
High cinder-contents are desirable for free welding, for such
purposes as pipe making, etc., and low cinder-contents when
the product is to be subject to severe physical strain.
The order and proportions of the oxidation of the principal,
metalloids are shown in the diagram, Fig. i.
The object of agitation, which is ordinarily accomplished by
the rabble, is, to produce as uniform conditions as possible
throughout the bath, to bring into intimate contact the cinder
and iron, and to prevent as far as possible the settling of the
iron, in a partly refined condition, on the relatively cold bot-
tom, and its becoming, as a consequence, too cold on the under
side. This occurs in a measure during the " drop," particu-
larly in large furnaces, and necessitates " turning."
Small heats, such as are still worked at Low Moor, England,
are best adapted for uniformity of product, since the workman
has not the physical strength and endurance to agitate the larger
baths efficiently, though he obtains material assistance from a
good '' high-boil." Another disadvantage in large furnaces is
the necessity of starting to draw " young," since otherwise
the later balls are subject to severe loss. The irregularities
thus produced are, however, largely overcome by the subse-
quent treatment of piling and reheating. Piling operates as a.
fairly efficient mixing by the natural law of chances; and re-
heating helps, by maintaining the iron at a temperature, and
for a x^eriod, which permits both the oxidation of the carbon of
the " young " iron and the reduction of some of the cinder
present. Hence, the small pieces of which an iron pile is
formed are desirable under existing methods of producing pud-
TJic Manitjaciitrc ajui CJiaractcristics of Wroiii^ht Iron 205
died iron. But piling has also its disadvantages, in the pro-
duction of laminations, due to carelessness, or lack of means
or skill, and m the high cost of handling loose piles.
IV. The Structure of Puddled Iron
By reason of decreased fusibility through the elimination of
carbon puddled iron crystallizes (" comes to nature ") at a lower
10
20
50
40 50 6U
TIME fMINUTES)
70
80
Removal of Impurities from Iron during the Puddling Process
temperature than steel. Each grain or crystal is surrounded
by an envelope of cinder, which, when the clusters are dense,
fills the inters^ening spaces. The greater part of this cinder is
then hammered or squeezed out, and the succeeding operation
of rolling elongates the crystals into what are commonly known
as fibers, each fiber existing in a matrix of cinder. This fibrous
formation in a matrix of ferrous silicate is the controlling char-
acteristic of wrought iron, and the source alike of its virtues
2o6 The Iron and Steel Magazine
and its faults. The fibrous structure can be distinctly seen un-
der the microscope at variotis powers, and often with the naked
eye. It is, perhaps, more correctly described by Professor Howe,
who says: " I understand that this ' sort of fiber ' is more ap-
parent than real, the grains themselves being equi-axed, yet
separated into quasi-fibers by layers of slag. ..." * Admit-
ting the correctness of this remark, each series of crystals forms
an integral structure, which has to be ruptured separately, pro-
ducing, in every-day ]:»ractice, results analogous to those we
would look for in a true fiber.
Many instances of the arrest of fracture by the fiber occur
in the experience of most engineers, with shafts, bolts, chain-
hooks, etc. An illustration recently came under m.y observa-
tion : About two years ago a shear cam-shaft, 9 inches in diameter
by 48 inches between journals, was bent about 0.5 inches at
one side of one of the two cams, producing a crack about 2 inches
deep and open almost an eighth of an inch. The lateral move-
ment of the large spur-wheel, due to the bend, was carefully-
noted, and measured at frequent intervals to learn whether the
fracture was extending. As it did not extend, and no conven-
ient opportunity offered for straightening, the shaft ran till
July of last year, when it was taken out, straightened and re-
placed. It is doing good work, and running true to-day. Had
it been steel, it would have been necessary to take it out at once,
and, after straightening, anneal it; even then, a greater risk
w^ould have been run than with wrought iron.
Drive pipes for wells furnish another illustration. Wrought
iron withstands the shock of driving at the threads, whereas
steel breaks off at the root of the threads.
A blast-furnace plant of two stacks had an experience bear-
ing upon this question. One stack was erected about thirty years
ago, with a shell of wrought-iron plates; the other, about four
vears ago, with a shell of basic open-hearth steel plates. Both
were evidently subjected to the same character of strains, and
in approximately the same position. The wrought-iron shell
bulged out under strain, but did not crack; while the steel
shell suffered a vertical rupture extending about 20 feet, and
not following to any marked degree the lines of the riveted
joints.
* " The Metallurgy of Steel," by Henry M. Howe. p. 1Q3.
TJic Maiiitfaciitrc atid CJhiriWtcristics of Wrou^Jit Iron 207
Physic^il tests of steel made in the laboratory on an 8- or
lo-ineh seetion i^i\'e results whieh superficially appear to be
siiperior to those of wroui^ht iron, particularly in the feature of
eloni^ation. But when studied with greater care the elonga-
tion is seen to be, largely concentrated, and not so uniformly
distriliuted throughout its length as is that of iron.
This tendency to concentrated elongation in steel, and to
more widely distributed elongation in wrought iron, is shown
in their respective tests, in full-sized sections, for eye-bars in
bridges, the specified elongation being the same for iron and
steel in a length of 10 feet. The following are results of such
tests:
El.istic Limit
Ultimate
Strength
Elongation,
per cent
Reduction
Area
of
In 12 in.
In 18 ft.
Per
cent
23
15.22
28
•30
39
14.40
51
•50
Iron
Steel
Lb. per sq. in.
31.550
33.150
Lb. per sq. in.
48,810
59,260
There is greater general confidence in welded articles made
of iron than of steel. This is well founded: the cinder present
in iron, and its low carbon content, naturally facilitate welding.
This receives further confirmation from the fact that, as the
carbon is increased in steel, the uncertainty of a good weld
becomes greater.
V. Resistance to Oxidation
That wrought iron resists oxidation better than steel is
becoming the general opinion of those who have studied the
question under actual w^orking conditions. The difference is
naturally more apparent in thin objects, such as corrugated
roof-sheeting, tin plate for roofing, and the like ; but its influ-
ence is the same regardless of mass. The difference in the life
of light sections is about as five to one in favor of puddled iron.
The explanation of this resistance to oxidation is twofold:
I. The cinder, a ferrous silicate, enveloping each fiber, is
much attenuated by rolling, and in that condition is elastic. A
piece of iron fresh from the rolls is covered with relatively thick
scale, which will readily crack off to a large extent, exposing a
2o8 The Iron and Steel Mamzine
i>^
•surface of iron fibers with its intervening cinder. These fibers
oxidize somewhat rapidly, leaving a finely corrugated surface of
cinder, which resists further atmospheric action, as may be seen
in heaps of scorioe from old hearths, believed to date from be-
fore the Christian era. Being elastic, it resists for considera-
ble though varying periods; but eventually it cracks off under
vibration, expansion and contraction, or mechanical wear. The
cycle is then repeated and so on.
2. Puddled iron is a mechanical combination of two sub-
stances, iron and cinder, which offer differing resistances to
such pressure as that of rolls or hammers. The result is a
rough surface, which forms a more lasting bond with any pro-
tecting agent, such as tin or paint, than the smooth surface of
steel, which does not aid in any way tlie adhesive qualities of
the protecting agent.
In connection with the question of oxidation, I may iUvStance
the experience of a large tube works, carrying a considerable
stock of iron tubes, and accustomed to take from and add to
the top of the stock-pile, without regard to the tubes in its lower
part, knowing that these, when ultimately reached, would be
found to be corroded tmiformly over their whole surface, but
could be re-rolled to a lighter gauge, producing perfect tubes.
After beginning to make steel tubes, they followed the same
practice; but these tubes were found after re-rolling to be
pitted through, and therefore valueless.
I am indebted to Dr. RaA.'mond, secretary of the Institute,
for the suggestion contained in the following communication:
" In preparing yoiir paper for the press, I notice that you
have omitted to mention, in connection with the question of the
more rapid oxidation of soft steel, a chemical reason, namely, the
presence of manganese in that metal. Many years ago, as con-
sulting engineer of the firm of Cooper, Hewitt & Co., I approved
the substitution of low-carbon steel for wrought iron, for certain
articles of manufacture. The immediate result was complaint
from both consumers and selling agents, that these articles rusted
so soon as to look old even upon delivery. A careful investiga-
tion, conducted for the firm by the late Dr. T. M. Drown, located
the source of this trouble in the manganese of the low steel, or
' ingot-iron.' In that particular case the rapid surface corro-
sion probably did not affect the real usefulness of the articles.
The Mamtfaciiirc and CJiaractcristics of WrougJit Iron 209
But it may easilv be inferred that, when a coating of tin, zinc
or paint is applied to a sheet of metal a very slight extra liability
to oxidation in that metal may set up a series of chemical and
galvanic reactions of destructive character.
" I have had recent occasion to realize with surprise and con-
sternation the imperative necessity of frequent repairs to roofs,
pipes, etc., of tinned or galvanized iron. My trusted mechanic
declares that all his customers are similarly affected, and pro-
tests that he can no longer obtain anywhere materials of this
class as durable as they used to be. He thinks that something
is the matter with the processes of coating with tin or zinc ; but
I shrewdly suspect that the trouble lies in the manganese of
the metal coated, and in the series of reactions which its easy
oxidation initiates.
" It seems to me that the ' pitting ' of steel, to which you
refer, is directly due to manganese."
VI. Defects of Wrought Iron
Under this head we have to consider transverse weakness,
lower ultimate strength, laminations and high labor cost in
production.
1. Transverse weakness is inherent, though less marked as
the cinder contents are reduced.
2. Lower ultimate strength and elastic limit, demanding (for
the same factor of safety, and if the possible effects of the " pit-
ting " of steel be ignored) a greater section for a given strain,
is also an inherent weakness, so long as the product is made
from built-up piles, requiring low carbon as an essential condi-
tion for good welding.
3. Laminations, due to imperfect welding, are commonly the
result of inefficient machinery or lack of skill. This defect
also is inherently associated with the use of built-up piles, es-
pecially when these are made, wholly or in part, from scrap of
miscellaneous character. I recall an instance in which a lot
of i|-inch " rounds " were ordered to be miade for special bolts,
from muck-bars only. While cutting them to length on the
anvil we noticed that, after superficial nicking, some pieces
fell off, even without bending. Upon investigation, the so-
called " iron " was found to include pieces of hard steel rail,
2IO The Iron and Steel Magazine
fairly well defined in form, and surrounded by soft wrought
iron, presumedly " muck-bar," which peeled off with the free-
dom of a banana-skin, the exterior and the core not being
welded together.
4. The high labor cost of production is probably the principal
cause which has checked, during recent years, the legitimate
increase in the use of wrought iron. It is constantly receiving
greater emphasis, as the younger generation of workmen, with
greater educational adv^antages, turns away from arduous man-
ual toil, and particularly from the opprobrium attached to the
term " puddler," as indicating a relatively servile and ignorant
class. This stigm^a has existed from time immemorial in India,
where the iron-worker is ranked in the lowest caste. The se-
verity of the work, together with the inferior social status of the
worker, have depopulated puddling mills in this countr}^ and
England, and caused, in certain districts, the abandonment of
puddling.
With the exception of relative transverse weakness, all the
above shortcomings of wrought iron are due to the method of
manufacture, rather than to inherent qualities of product. Iron
puddled in large miasses by mechanical means, and rolled direct,
as soft steel is, would give us fibrous structure, resistance to
oxidation, high tensile strength, low cost, and the absence
of laminations. Such a material w^ould cover the field now
jointly held by puddled iron and soft steel, even to the possible
use of the latter for rails.
VII. Conditions Essential to Successful Puddling
The conditions necessar}^ to produce such iron are: (i) a
large unit of manufacture; (2) adequate mechanical means;
(3) cinder of proper composition; (4) a flame of the right com-
position and temperature; (5) a relatively permanent furnace
lining; (6) a relatively small loss of iron; (7) simplicity of
means and method.
I and 2. Conditions (i) and (2) are the chief factors (in-
deed, if we consider Bessemer steel alone, the only ones) in the
low cost of steel manufacture. These two conditions, intro-
duced into the production of wrought iron, would have a like
effect upon its cost. I will consider later the means of effect-
ing this end.
77/t^ Mauiifaciure and Characteristics of Wrought Iron 211
3. As already shown, the right cinder for the i uddHnc: pro-
cess is largely produced from, and corrected by, the oxides
forming the sides and bottom of the hearth. It is. how-
ever, evident that the purer oxides of iron are inert while in
position, and only become active after absorption into the bath.
Hence, when suitable oxides are introduced in the form of cinder,
ore, and roll-scale, there is no need of drawing upon the bottom
and sides, provided these latter are of such nature as to resist the
chemical action of the cinder and the temperature of the gases.
4. The proper regulation of the flame is essentially a simple
matter, presenting difificulty only when associated with the
motion of the furnace (if the latter be movable).
5. The lining offers a somewhat complex problem, both
metallurgically and mechanically. It must resist the chemical
action of a compound cinder, the friction of the mass as the
iron comes to nature, and a somewhat high temperature.
Moreover, in a mechanical furnace, it must maintain its posi-
tion throughout the movements of the furnace itself.
6. The minimum loss of iron is secured by the reduction of
some of the iron oxide additions, resulting in an actual gain of
weight of the puddled mass, over the pig charged, in an ordi-
nary puddling furnace. This gain is exceeded by the final loss,
due to the delay in the period of balling and drawing. By
reducing the period of drawing, as by discharging the whole
mass in one piece, this loss may be avoided, and a possible net
gain effected.
A mechanical puddler, erected at Pottstown, Pa., and fulfill-
ing the conditions above stated, was described by the writer at
a former meeting of the Institute.* To that description, the
reader of the present paper is referred.
* " Puddled Iron and Mechanical Means for its Production," Phila-
delphia meeting, May, 1892, Trans., XXXIII, 551.
212
The Iron and Steel Magazine
THE THERMIT PROCESS IN AMERICAN PRACTICE *
By ERNEST STUTZ
JUST a year ago the first Thermit was manufactured in this
country and tlie applications developed in Euro])e by Dr.
Hans Goldschmidt, at the works of Th. Goldschmidt, Essen-
Ruhr (founded 1847), were transplanted to American soil and
have since blossomed forth under the fostering care of American
ingenuity.
The principle of the Thermit process can now be said to
be known to the technical world, and it will be sufficient to state
that through the ignition of finely divided aluminum and
metallic oxide a reaction is started which produces heat at
about 5400° F. and at the same time reduces the iron oxide to
a metallic iron almost free from carbon, in a highly superheated
liquid state. Thermit, steel has practically twice the tempera-
ture of open-hearth steel, and a correspondingly greater fluidity.
By suitable additions of carbon, in the form of steel punchings,
chilled iron shot or ferro-silicon, its hardness, and, Vjy addition
of manganese, its toughness, can be increased to any suitable
degree.
The following analyses will confirm this:
ANALYSLS OF TIIERMir STEEL
Illinois Steel Company, the Rookery, Chicago, III.
Carbon
0.05
Manganese
O.IO
Silicon
0 204
Sulphur
0.04
Phosphorus Aluminum
0.05 0.18
i
Tensile Strength, lbs. per sq. in.
59»320
Elongation
25.33 per cent
Contraction of Area
59.6 per cent
Pennsylvania Railroad, Alioona
Thermit Steel with addition of 2 per cent carbonless manganese 5 per cent iron punchings
(calculated on amount of Thermit)
Carbon
Manganese
2.330
Silicon
1.227
Sulphur
0.034
Phosphorus
0.070
Aluminum
Tensile Strength, lbs. per sq. in.
9 1 ,600
Elongation, per cent in 8 in.
21-5
Appearance of Fracture
Silky
■
* Read at the June, 1905, meeting of the American Society for Test-
ing Materials.
77/l' Tlicrniii Process in Anicriaui Praciicc
213
The tirst is one of pure Thermit steel ; the other of the steel
in the riser of a welded-steel locomotive frame, drawn out under
the hammer into a bar some three feet long and turned down
and broken.
The simplicity of outfit and manipulation and the speed
with which the reaction does its work are its chief recommen-
dations for industrial purposes.
In a crucible some 20 inches high and therefore easily trans-
portable, in half a minute can be produced 30 pounds of liquid
steel, so hot that it will melt a steel bar of 4 inches square section
and fuse with it to one homogeneous mass.
The essential characteristic of Thermit is that it welds by
fusion, and, by reason of this fact, calls for the foundry man's
experience more than the blacksmith's. Its success depends
on the proper material, shape and condition of the mold.
The mold into which the contents of the crucible are run
must be of refractory material. The general instructions must,
of course, be broad and cannot go beyond stating that a mixture
of equal parts of sharp sand and ordinary brickmaker's clav has
given satisfaction. The formula has been varied sometimes,
according to local conditions, in some cases flour, in the pro-
portion of 6 to 100, being used as binder for the sand. Some
shops have already evolved their own particular formulas, which
they treat as secret. The mold always must be dry — burnt
dr}-. In some cases, for instance, at the Elkhart shops of the
Lake Shore & Michigan Southern, the difficulty has been over-
come by using firebrick cut down to size. This certainlv over-
comes the question of drying molds.
The shape of the mold must next be considered. It must
be so constructed that the steel flowing down through the gate
will not strike direct on to the casting or forging, but will flow
underneath the lowest part and rise around and through it.
What is required is good circulation for the Thermit steel. It
must flow around all the welding surfaces, and as it gets chilled
in contact with these it must be driven up into a riser and be
followed by a sufficient supply of fully heated Thermit steel to
effect the actual weld, which takes the shape of a collar or rein-
forcement, cast on or over the fracture.
The mold mu.st, therefore, allow (i) for a gate; (2) for a
collar, shoe or other reinforcement on the surface of the welded
21
The Iron and Steel Magazine
piece and overlapping the edges of the break or joint; (3) a
riser; (4) a skim gate, to prevent the slag from getting mixed
with the steel.
The formula for calculating the amcuit of Thermit must
also allow not only for the cubic space of this reinforcement,
but further, for again as m^uch Thermit, to supply the contents
of gate and riser.
These are the general instructions for welding, for instance,
locomotive frames — a problem which some thirty railroads in
this countrv have inves^i<^ated with more or less success. These
'^^^^^
Break of Welded Bar, 2I x 2f, after Pressure of 50 Tons
frames are of wrought iron or cast steel and vary from i\ x. 3^-
to 5 X 6 inches in section. They are very liable to break and
their repair without dismantling the engine means a very large
saving per engine. It has been -stated that an engine the frame
of which is repaired in the forge remains a fortnight out of com-
mission and the actual weld costs $250 to $300. The work by
Thermit can be done comfortably in three or four days, at a
cost of about $50.
In reply to a circular letter of inquiry, about twenty rail-
roads have supplied data, which, however, cannot be considered
The TJicmiil Process in ADuiican l^ractice
215
complete, as some of the most reg^ular and extensive users of
Thermit did not care to supply the information asked for.
The first successful weld it has been possible to get a record
of was made by Mr. Sanderson, suj)erintendent motive power,
Seaboard Air Line, on October 19, 1904. This engine has con-
tinued in service ever since. It is one of eight engines welded
on that road which has given satisfaction, which speaks highly
for the care used at the Portsmouth shops in handling a new
and therefore difficult problem.
Another series of successful welds is reported by the Boston
Welding Locomotive Frame: Ready for Ignition
& Albany Line, where Mr. Fries welded five engines quite suc-
cessfully, one being in continuous service since the end of
Xo\'emV;er. One. welded in the jaw, broke again, but four
inches away from the weld.
Of late the Lake Shore & Michigan Southern has shown great
interest, and its perseverance has been crowned b}^ success in
some very good welds at their Elkhart shops, about which Mr.
WebV) read a very interesting paper at the last annual meeting
of the American Foundrymen's Association, giving a full account
of each step in the operation. On a preliminary test, a welded
bar 2J X 2^ stood a pressure of 50 tons on supports 20 inches
2t6
The Iron and Steel Magazine
apart, before breaking, and that after two sides of the reinforc-
ing collar had been machined off.
In all there are records of thirty engines with welded frames
that have been in service for three months or longer. Failures
are recorded only in isolated instances and are assignable to
two different reasons:
First, wrong construction of mold.
Locomotive Frame: Welded in the Jaw
Second, insufficient Thermit; in other words, insufficient
circulation — therefore, insufficient fusion.
For those familiar with the process, a weld that breaks on
account of lack of cohesion at the welding surface is attributable
under all circumstances to lack of experience or care, except in
one particular case.
It is possible for Thermit welded frames to break in spite
of proper execution of the work. The original break is due, in
the first place, to a structural defect. With the l^reak in such
The TlhiDiii Process iii Aiiicricaii Pvacticc
217
a i)osition as to necessitate the entire removal of the reinforcing
collar, it is too much to expect the mere bridging of the broken
ends In' Thermit steel to overcome this innate weakness.
An important factor in success in welding locomotive frames
is to allow for equal shrinkage of parallel parts; also, wherever
possible, to spread the ends apart in order to let them come back
when the iron begins to set.
Another operation of interest to railroad men is the welding
of spokes of drivers.
In making tests of the metal of such welds, the Chicago,
Welding Spoke of Locomotive Driving Wheel
Milwaukee 8c vSt. Paul R. R. found a tensile strength of 93,900
pounds per square inch. The analysis agreed with that of the
Pennsylvania R. R., with the exception of manganese, which in
this case was only 0.74.
Next came repairs in marine engineering, which are mostly
successes obtained by Mr. Des Angcs, superintendent floating
equipment of the Long Island Railroad.
A 12-inch crank shaft (13 g inches at point of fracture) of
the fern^-boat Manhattan Beach was welded with 400 pounds
of Themn't. The break was in the " wheel center," necessitating
2X8
The Iron and Steel Magazine
the shifting of the center to a new j^osition and shortening the
paddle boxes. The shaft was pre-heated, by a charcoal fire and
hand-blower, to black heat. To protect the woodwork of the
ferrv-boat an asbestos ctirtain was hung around the crucible,
which served its purpose admirably. The ferry-boat has been
in uninterrui)ted service for nearly three months, and continues
so now.
A rudder stock 5 inches in diameter was welded with 50
pounds of Thermit and 10 pounds of punchings. The collar in
this case had to be entirely removed, but the welded rudder-
stock has now been m service for eight months.
iVsiDSO Shaft
Weld of Crank Shaft, Manhattan Beach
On the Great Lakes, through the enterprise of Captain
Johnson, at that time with the Dunham Towing ^' Wrecking Com-
pany, the rudder shoe of the tugboat Sckenck was welded, 125
pounds of Thermat being used. The weld was sound — in re]- lac-
ing the propeller a chain broke and the propeller dropped on the
welded shoe without injuring it.
Some im];ortant repairs of gray iron castings are also re-
ported. At the Renovo shops of the Pennsylvania Railroad
a hydraulic wheel press was repaired, the part welded having
to stand a pressure of 60 tons per square inch. The origuial
" strong back " holding the wheel against which the axle was
77/r llicruiii Process in Aiucrican Practice
219
pressed was not stroni]: cnoiuj:h for the jnirpose until repaired
by Thermit.
Cyhnder eovers are also repaired by Thermit and have
been made as good as new.
Work with gray iron castings requires more experience, in
regard to pre-heating and cooling down gradually — more
Thermit is necessary to effect the weld, on account of a hard,
<:ylassv scale on such castings, which resists fusion, and an addi-
Weld of q-in. Rudder Stock
tion oi fern^-silicon (about 2%) is advisable to prevent hard
spots at the lines of junction between Thermit steel and cast iron.
The most important application of the Thermit process is
for making a continuous rail. The process having been brought
to a high state of perfection in Europe before coming here, there
was little room for changes in practice. About 30 different
cities are investigating the process in actual operation and about
5,000 joints have been put in up to date. All the.se roads recog-
220
The Iron and Steel Magazine
nize in the Thermit process the best and simplest means of join-
ing rails for electric traction, as long as care is taken to do small
and sim.ple things right. Competitors in the field of rail welding
may send out fanciful blue-prints about broken joints, to create
unfavorable impressions, but such man oeu vers prove nothing
beyond the fact that they admit the success of the Thermit pro-
cess in this field.
Some tests may be of interest. A heavy double trolley
car was taken over a welded joint with supports 13 feet away,
without breaking it.
Welded Rudder Shoe, Tugboat Schenck
To decide whether the head of the rail got softer, microm-
eter caliper measurements were taken of depressions made under
equal blows of a steam hammer, by a blunt tool hardened at the
head, J inch in diameter.
One-half inch away from the joint the depression was 0.1432
inches.
Three feet away from the joint the depression was 0.1596
inches.
The electric conductivity of the Thermit joint is recognized
to be higher than that of the rail, due to increase of area, and is
permanent.
77/1- llicrmit Process in American Practice
221
That steel foundries should liave been the first to recognize
the possibilities of liquid steel that can be produced anywhere
in half a minute goes without saying. There are already several
of the largest with whom Thermit is as much a necessity as foun-
dry sand. Some prefer — for no apparent reason — not to
disclose the fact that they repair faults in castings by Thermit,
but all can openly admit that they use it to reduce the size of
their risers, an application which, through its simplicity, recom-
mends itself to all foundries, — gray iron as well as steel. Ther-
mit thrown loosely or in a paper parcel on steel will ignite and
Welding Trolley Rail at Holyoke, Mass.
keep the contents of the riser fluid even after the metal has
become plastic in the casting. Liquid cast iron will only ignite
Thermit in the presence of the ignition powder.
The application of Thermit to reduce the piping in ingots,
although very simple in itself, necessitates some liquid steel
being held in readiness to fill up the piping after the solidifica-
tion has been interru]jted by a thermit reaction. This should
not be impossible to arrange.
222
The Iron and Steel Magazine
RiEHLE Bros., Testing Machine Company
TESTS ON MALLEABLE IRON BARS CAST AT PENNSYLVANIA MAL-
LEABLE company's WORKS, McKEES ROCKS, PA.
Before Titanium Thermit Reaction
No. I —
Dimensions
Ultimate Strength,
Poimds
Deflection
I — I
i.ooo X .999
4,100
1. 00"
I 2
■995 X .999
4,500
.98"
1—3
Lost in anneal
2—7
1.060 X .998
4,540
1.28"
2—8
1. 012 X 1.006
4,610
1.40"
2—9
1.006 X 1 .005
Average
4,500
1.40"
before treatment
4,450
1. 212"
After Titanium Thermit Reaction
No.
Dimensions
Ultimate Strength,
Pounds
Deflection
lA- 4
1. 01 I X 1. 010
5,920
■ I ■30"
lA- 5
.999 X 1 .000
4,260
1.27"
lA— 6
.989 X .995
4,850
155"
2 A — 10
■995 X .996
4,620
1.47"
2 A — 1 1
.998 X .996
4,410
1^3 7"
2 A — 12
1. 01 1 X I.ooo
Average
4,810
1.44"
after treatment
4,811
1.60"
Another branch of alumino-thermics which will be of inter-
est is the improvement of gray iron castings, by the introduction
of titanium Thermit in the ladles, by immersing it in a cartridge
below the surface of the metal. Some experiments, thanks to
our fellow-member's, Dr. Moldenke's, kind intercession, were
made at the Pennsylvania Malleable Works, with the foregoing
results, the bars having been poured out of the same ladles, one
before, the other after, the titanium Thermit reaction.
Experiments witli lower grades of iron showed the same
favorable results.
At the Featherstonc Foundry, Chicago, titan Thermit
treated test bars showed a tensile strength of 3,550 pounds,
against average untreated, 3,250 pounds.
The metal, after treating, is much denser, but can be easily
machined.
The Thermit Process in America]! Tracticc
223
Incidentally it may l)e n-icntioncd that by the introduction
of a lA-pound cartridge of ordinary black Thermit into an 8co-
pound ladle 40 pounds of steel borings can be melted without
difficulty.
This necessarily very short account of what is doing in
Thermit cannot, of course, cover the entire field of the applica-
Welding Rudder Stock on Tugboat Schenck on Marine
Railway, Sault Ste. Marie
tions, but will perhaps tend to convince those who had rather
be guided by results obtained elsewhere than spend time and
money for what they think experiments, and encourage others
who are doubtful from lack of ex])eripnce, by showing them what
has been accomplished in actual practice.
2 24 The Iron and Steel Magazine
THE APPLICATION OF DRY-AIR BLAST TO THE MANU-
FACTURE OF IRON *
By T. W. ROBINSON
'npHE international discussion that has been evoked by Mr.
-'' Gayley's latest contribution to the annals of iron and steel
is a striking testimonial to the value of his paper. If the fig-
ures given can be considered typical of what may be reasonably
exx)ected from the use of the dry blast, its innovation is bound
to prove of the greatest importance. That the problems of cause
and effect are intensely interesting, both from their commercial
and scientific aspect, is reflected in the widespread interest
that has been evinced. Those questioning the sincerity of the
Isabella test, or the accuracies of recording the results, will not
be considered by those acquainted with Mr. Gayley or his past
work. As to the causes for results so at variance with accepted
ideas, the discussion has brought out little that is convincing.
It is strange that greater stress has not been placed on what
Mr. Gayley, himself, has had to offer in the way of explanation.
Stated briefly, his reasons for the higher efficiency obtained
are, greater uniformity of practice and ability to reduce the
heat reserve. The essence, of course, of the Isabella demon-
stration is that the furnace, working under as strictly compara-
tive conditions as could be devised, showed a sa^•ing of approx-
imately 400 pounds of coke per ton of iron through the substitu-
tion of dry blast for ordinary blast. The natural inference was
that this result must be directly due to the saving of the heat
ordinarily dissipated by the usual introduction of the moisture.
Calculation, however, shows that the heat units involved in the
evaporation and disassociation of the moisture removed by
refrigeration represent less than one fourth of the amotmt of the
fuel actually saved. How can such an anomaly be explained?
In studying this question one naturally looks, first, to the con-
ditions under which the test was made. We have Mr. Gay-
ley's assurance that every care was exercised to make the
conditions under which the furnace was operated strictly com-
parative. The records at Isabella, which were thrown open to
* Discussions of the paper of Mr. Gayley read by title at the Lake
Superior meeting, but first presented at the New York meeting of the
Iron and Steel Institute (October, 1904).
Applicatioji oj Dry-Air Blast to the Manufacture of froji 225
inspection, verified this fact. Raw material, equipment and
general practice were sliown to be practically constant during
the test. The only appreciable variable was the character of
the blast. Mr. Gayley's papei covered an operation of one fur-
nace for onlv about six weeks, during which the fuel was re-
duced approximately 400 pounds. It has been questioned
whether so short a test might not easily lead to erroneous con-
clusions. Fortunately for the advocates of the dry -blast process,
further experiments at Isabella have corroborated the initial
returns. Since the paper w^as presented the writer has examined
the records of four consecuti\'e months on dry blast at the two
adjacent Isabella stacks. During this period both were operated
under practically identical conditions save that one was run
with drv blast and the other with ordinary blast. The dry-
])last furnace showed an average coke consumption of 437
potmds per ton of iron less than the other. To make the experi-
ment more conclusive the blast on the two stacks was inter-
changed, the one formerly on dry air being put on ordinary air
and rice rersa, the practice otherwise remaining the same. In
less than a week the furnaces had exchanged places in production
and fuel consumption. This further demonstrated that any
differences that might have existed in plant and practice were
not of much consequence. The logical conclusion is that the
refrigeration of the blast must, in some manner, be accountable
for the surprising results obtained.
The problem, accordingly, resolves itself — (i) into a study
of the effect of the refrigerating process on the character of the
blast itself and (2) of the influence of the dry l)last on the fur-
nace reaction.
Refrigeration as carried out at Isabella eliminated about
three quarters of the normal moisture and materially increased
the density of the blast at the blowing tubs. The fuel directly
saved by the elimination of the moisture, as based upon Mr.
Gayley's figures of 69 pounds of water removed per ton of iron,
is equivalent to less than 80 ]jounds of coke. If we add to this
the saving incident to the increased temperature of the blast
there is still a large ])art of the total saving unexplained.
But what effect has the increased density? The records
show that, where formerly the engines were making 114 revo-
lutions, the refrigerating of the blast necessitated a reduction
2 26 The Iron and Steel Magazine
to 96 revolutions. If, to be conservative, we assume the ordi-
nary temperature of the air to be 75° F. and that of the refrig-
erated blast, as it enters the Ijlowing tubs, to be 22° F., and
allow the refrigerated blast to have an increased pressure of
0.3 oiuices, due to the use of the auxiliary fan, we find that
the actual weight of the air used per unit of time is in the ratio
of IOC to 93.6. In other words: By slowing down the engines
from 114 re^■olutions to 96 revolutions there was a reduction
of a little over 6 per cent in actual weight of air blown. In
volume the reduction at the blowing tubs amounted to 15.8
per cent. This, however, under equal furnace pressure, would
have been reduced to a difference in volume of 6 per cent at
the tuyeres had the stoves heated the air, in each case, to a
like temperature. While weight and volume did not corre-
spond at the engines they would haA^e done so at the furnace
had the stove capacity been sufficient. That the stoves did
not, in each instance, heat the blast to the same temperature is
merelv incidental and not essential to the case in point.
Accordingly, as increasing the density at the blowing tubs
does not change the volume of air, or the weight of air, or
the weight of oxygen that enters the furnace, if compensated
for bv lower engine speed, its only appreciable effect is to in-
crease the engine efificienc}^ The more dense the air the
greater the weight that can be blown per revolution, that is
all. Hence it is clear that the increased density of blast, due to
refrigeration, cannot rationally be offered as a cause for the
decreased coke consumption. Further, the only practical result
in cutting the revolutions from 114 to 96 was to reduce the
bla.st from 40,000 cubic feet per minute to about 37,400 cubic
feet per minute. This is a comparatively small reduction and
one, of course, capable of attainment without the intervention
of refrigerating machinery.
Mr. Gaylev says, that by reducing the revolutions from 114
to 96 the volume of blast was reduced 6,000 cubic feet per minute.
He cites this fact in calling attention to the influence that the
increased density of the air had upon the steam economy. His
statement is correct, but liable to misconstruction imless the
material difference in the relation of weight and volume in
ordinary and refrigerated temperatures is appreciated. Tlicy
are, of course, not comparable.
Applicatioii oj Dry-Air Blast to the Mauii]a<.titrc oj Iron 227
As usually calculated by piston displacement, the difference
in the dry and the ordinary blast that entered the furnace was
2,600 cubic feet, not 6,000 cubic feet. Apparently, then, neither
increased density nor reduction of blast could explain the balance
of the fuel still unaccounted for.
Turning now to the effect of the dry blast on the fiirnace,
we know that its introduction was quickly followed by increased
production and decreased fuel consumption. Analyzing the
details, it is perceived that the furnace on dry blast ran with
lower top temperature, with an increased per cent of carbonic
acid in the waste gases, with a less amount of air per unit of
coke and with lower pressure of blast. The differences are
indicative of lower fuel and are the results of less coke consump-
tion, not causes for it. We note that the iron produced is more
\miform in analysis, that there is less slipping of the furnace,
that the flue dust has been reduced from 5 per cent to i per
cent and that, generally, the dry -blast furnace is more regular
in its operation than the other. In studying the y)ossible
causes for fuel saving, it has been shown that the difference
existing in plant and raw material had no appreciable effect.
The influence of increased density and decreased revolutions
of engines was practically negative. Allowance for the heat
equivalent of the eliminated moisture and for the increased
temperature of blast does not explain the results that were ob-
tained. One point, however, that has not yet been discussed
is, the greater regularity of the dry -blast furnace. That greater
regularity ought to follow refrigeration will be apparent, if the
large amount of moisture usually introduced is appreciated and
the extent of its hourly and daily variations fully realized. If
any one questions this, let him consider what the effect would
be of injecting at the tuyeres a stream of water under a 4c-foot
head that varies from 0.75 to 1.5 inches in diameter. This is
practically what a humidity varying from 2 to 6 grains per
cubic foot means when 40,000 cubic feet of air is used per minute,
even though introduced as super-heated steam.
By reducing the moisture two thirds or more, and by causing
the residue to become nearly constant, the refrigerative process
cannot help Vmt have an important effect upon regular work.
Xow, it may be considered axiomatic that under any given
condition of raw material and equipment minimum fuel will
2 28 The Iron and Steel Magazine
accompany niaximiiin regularity. It is no uncommon occur-
rence to see a difference of 400 pounds of coke per ton of iron
follow the swinging of a furnace from uniform to irregular work.
Ordinarily irregular work is a direct sequence to scaffolding,
even though it be embryonic. As to the cause of scaffolding,
its name is legion. Lack of uniformity in any of the elements
of blast-furnace practice is conducive to a changeable zone of
fusion and to accretion. A change of 50° F. in stove temper-
ature will often raise the pressure and sometimes cause more
serious trouble. Largely to the extreme variation of moisture
in the outer air and in the engine room is, undoubtedly, attrib-
utable the fact that there is not a greater difference in fuel be-
tween summer and winter. A compilation of the average
monthly results for ten 3xars of eight furnaces at South Chi-
cago shows that for June, July and August there was an increased
fuel consumption of but 21 pounds of coke per ton of iron over
the winter months. As these furnaces were working, during
this long period, under conditions that were practically com-
parative, the result is of more than ordinary interest on account
of so manv variables being minimized. The possibility of
maintaining more uniform conditions through protecting the
incoming blast from steam contamination appears worthy of
more careful consideration than is ordinarily given.
The blast furnace is an exceedingly rough piece of mechan-
ism, but, withal, a most delicately poised apparatus, when work-
ing under greatest efficiency. When normal, its chemical and
physical laws are subject to reasonably accurate deductions;
but, unfortunately, its regular condition is one of irregularity.
This is paradoxical, but just here lies the main reason why Mr.
Gayle}" is obtaining something in practice that theory does not
fully explain. From a laboratory standpoint, the reactions
involved in iron smelting are simple and well understood. As
actually encountered in the furnace they are exceedingly complex
and difficult to measure — especially so when there is irregularity.
The question of removing the moisture from the blast is, of
course, an old story. Many have considered the subject and
discarded it as an unfeasible proposition, not only for want of a
proper method but from failing to recognize its latent possibili-
ties upon more uniform practice. Twenty years ago the writer
was cognizant of comparative tests simultaneously made by
Application of Dry-Air Blast to the Manufacture of Iron 229
Messrs. James Gayley, E. C. Potter and F. A. Emmerton, at
Pittsburg. Chicago and Joliet, upon the humidity of the air
and its effect upon the furnace. That, after so long a study,
Mr. Gavlev has been able to make the Isabella demonstration,
despite practical and scientific discouragement, is a fitting cli-
m:ix to his persistency and ingenuity. Theoretically, the end
did not warrant the means. But, in iron smelting, theory has
had to be remodeled many times to square with actual results.
Numerous reasons and hypotheses have V^een advanced in the
past to prove why it would be impossible to obtain certain ulti-
mates in coke and product, which were later achieved and sur-
passed, and history re|)eats itself. The essential reason for the
more effective work that followed the replacing of old-time
methods by scientific management was the ability to command
more uniform conditions. In later years the evolution of fur-
nace practice has proceeded along divergent lines. On the one
hand, more perfect micchanical appliances and a better imder-
standing of the principles involved have tended to greater uni-
formity and lower fuel. On the other, much larger units,
heavier blowing and, in this country, the advent of the finer
ores from the Mesabi Range, have prom^oted irregularity and
higher fuel.
Because the Isabella furnace was consuming 2,147 pounds
of coke before the dry blast was applied, the question has been
raised whether the subsequent saving of 421 pounds is not mis-
leading, on account of being based on poor practice. Those
who know the facts are aware that 2,150 pounds of coke per
ton of iron more nearly represent the average practice to-day in
the United States than 1,900 pounds or any less. It is self-
evident that a furnace consuming 1,900 poimds would not pre-
sent the same opportunities for reduction as would one using
2.147 ])ounds. But the opinion that the Isabella practice was
poor is neither warranted by the conditions that existed nor by
the results that are generally being oV)tained in this country.
Whether the Isabella results can be considered a fair index of
what the dry blast would accomplish elsewhere will depend upon
individual conditions. It is certain, however, that the refrig-
eration process will ordinarily pay a handsome return on tlie
investment, and indications now point to its being widely intro-
duced.
230 The Iron and Steel Magazine
As to the influence of heat reserve, to which Mr. Gayley
calls attention, every practical furnace man knows how exceed-
ingly difhcult it is to correcth^ measure the actual reserve that
is being carried at any given time. In the Isabella case an ex-
amination of the records shows that, during the four or five
months of cold-blast practice, whatever reserve was carried lay
rather in the burden than in the stoves. As the amount of
reserve heat is best measiired by the analysis of the iron that is
being produced, it is significant that, during the period mem-
tioned, the iron from the furnace with ordinary blast was no
hotter than that from the furnace with dry blast. There was
more variation in the percentage of silicon, but the average
indicated that the maximum possible burden was being main-
tained on one as on the other. Accordingly, while the ques-
tion of heat reserve might be an important one, the evidence
fails to show that it was much of a factor in the Isabella test.
From what has^gone before, the logical conclusion is, that
greater regularity superinduced by refrigeration is largely the
cause of the extraordinary results at Isabella. This ma\ not
be capable of scientific demonstration, but until the usual
reactions of the blast furnace can be more accurately measured
I am content to rest rather upon actual results and rational con-
clusions than upon scientific deductions that are antagonistic to
facts.
THE INFLUENCE OF TITANIUM ON PIG IRON AND STEEL =^-
By PIERRE DELVILLE
1. Titanijerous Ores. — The ores contain from 10 to 40
per cent of titanium. They are found in imxmense quantities
in Sweden, Norway, Canada and the United States; they are
generally exceedingly free from phosphorus and sulphur, and
the percentage of iron ranges from 35 per cent to over 60 per
cent.
2. Direct Smelting of the Ores. - — The use of titaniferous
ores iti the blast furnace has been studied by Rossi, who contro-
verts the opinion that the presence of titanic oxide (TiOJ in
slag renders it more infusible and entails a higher consumption
* Presented at the Mining and Metallurgical Congress, Liege. " The
Iron and Coal Trades Review," July 7, 1905.
TJic liifliiciicc of Titaniurn on Pig Iron and Steel 231
of fuel. He also holds that the \ng iron, which usually contains
very little titanium (traces, to i per cent) is of good quality, and
equally suitable for the Bessemer process and for use in the
foundry. Some authors attribute this to the presence of tita-
nium, others to the low phosphortis, and others, again, to the
influence of slags containing titanic oxide, in eliminating sulphur.
The metal can be used for the subsequent production of armor-
plate and for V^oilcr plates, etc.
Titanium has been held to favor the formation of Vjears
and scaffoldins:, which are miore likelv due to an excess of lime
in the slags. Howe states that cyanides of titanium are formed
in the masonry of the hearth, but analyses of the bricks have
shown them to consist of almost pure lime. According to Rossi,
titanic oxide is totally irreducible by carbon, but it would appear
that the highest temperatures of the reducing and fusion zones
play an important part in this reduction, and pig iron has been
obtained containing 0.07 per cent of titanium, with a blast at
850° C, and a good coke. In this instance the burden yielded
about ;^T, per cent, with 10 per cent of Swedish ore containing
0.42 per cent of titanic oxide, so that a burden of 3 tons was
required to yield i ton of pig iron. With a hot blast the whole
of the titanium passed into the pig iron; if the blast became
cooler, and, in particular, if it fell to 700° C, the resulting metal
contained traces only.
3. Titaniferous Pig Iron. — In what state is titanium pres-
ent in pig iron? Rilev held, in 1872 and in 1880, that, in gray
pig, it was often present as a carbide. The presence of graphite
is not, however, necessary. White iron, high in carbon, may
contain large amounts. Thus, Allen cites an instance of a titani-
ferous plate containing:
Per Cent
Combined carbon 2.70
Graphite Traces
Silicon 1 .0 ;
Titanium 4.15
Manganese 1.37
Phosphorus 0.08
Sulphur 0.09
Shinier (quoted by Howe) says that the titanium takes the
form of little cubes of titanium carbide, insoluV)le in hvdro-
23 2 TJie Iron and Steel Magazine
chloric acid and in potassium hydrate, but soluble in nitric .
acid, and non-magnetic' Riley, at the miceting of the Iron and
Steel Institute in 1894, reported that the smelting of a Canadian
mineral containing i,T, to 34 per cent of TiO.^ yielded a pig iron
of splendid quality, but costing too much by reason of the enor-
mous quantity of slag produced (six times the weight of the
pig iron produced). At the same meeting, Dellwik explained
the desulphurization and possibly the dephosphorization of the
make, as due to the action of the titanic oxide, but the large pro-
duction of slag eat ailed an increased fuel consumption.
Bowron, the chemist of the Norton blast furnaces, which
made a specialty of the reduction of these minerals, stated that
the pig iron produced cost double that ordinarily produced, and
served for mixtures for imparting strength to foundry iron and
for open -hearth steel. Similar results were obtained at the
Adirondacks blast furnaces where, for years, the pig iron has
been used for the production of hard tool steels, samples of
which were exhibited in London as far back as 185 1. In 1892,
Raymond recommended the use of such pig irons in admixture
with ordinary foundry iron, for tires of wagons and for making
chilled castings.
4. Fer TO -Titanium. — The Aktieselskabet Titania Com-
pany, of Christiania, which owns deposits of the minerals, has
smelted them in the electric furnace and proposes to obtain by this
method a moderately fusible alloy containing m.anganese and
silicon or aluminum, in addition to the iron and titanium. The
use of this alloy appears simpler, although probably more costly
than that of Goldschmidt, hereinafter described (aluminothermic
process). Rossi has also jjrepared a series of alloys in the electric
furnace containing 10 to 75 per cent of titanium, and has im-
proved the quality of hard crucible steels by adding titanium,
thus increasing their malleability and ductility. These results
are probably due, as will be seen later, to the influence of tita-
nium as a deoxidizinga gent in the bath of metal, and to the elimi-
nation of the nitrogen occluded in the steel, the €X|ul£icn cf
which is favored by the presence of high percentages of carbon.
The" experiments, repeated by Rossi, gaw^ such promising results
th^t a company was formed to establish works at Niagara for the
preparation of these alloys under his patents. The results have
not yet been made public.
TJic hijlitciicc of Tiiaiiiiini oji I^ig Iron oiiJ Sicei 233
A French firni has produced a ferro-titanium containing
about 45 per cent of titanium by reducing a mixture of rutile
(TiOo"^ and iron oxide by means of aluminum, at a red heat,
in the presence of boric acid, in a crucible. Paul Girod has
given the author some particulars respecting the preparation of
ferro-titaniums in the electric furnace. The alloys contain
about 50 per cent of titanium and 47 per cent of iron.
The efifect of titanium on pig iron and steel is verv little
known. Analysis has shown it to be very difficult to recover
the titanium added, and its action has consequently been re-
garded as purely chemical; the absorption of gases contained
in the steel. Cast steel thus treated possesses, however, a great
surface hardness, resembling that conferred on quenching, and
this can be further accentuated by actually carrying out the
latter operation. Several crucible steel makers have, during
the past year, used these alloys for the production of self-harden-
ing tool steels, the effect being to increase the surface hardness
of these steels, while at the same time diminishing their brit-
tleness. The alloy is very refractory, and becomes with difficulty
fluid at the temperature of the electric furnace (3000° to 3500°
C). To get it to mix with the steel in the crucible, it must first
be heated to a white heat, and then thrown in while the steel
is very fluid, so as to make a homogeneous mixture.
5. Titanijerous Steel. — Hadfield considers that not more
than 0.3 per cent of titanium should be present in steel, the
proportion below w^hich it appears to absorb gas. It is to be
remarked, however, that under these conditions a cyanide of
titanium is found, which separates from the metal. Guillet, who
has studied these steels, has come to the conclusion that the
influence of titanium is practically nil, and that titanium steels
possess no commercial value.
6. The Use of A luminothermite. — Dr. Goldschmidt, of
Essen, makes a ferro-titanium with 20 to 25 per cent of titanium,
which mixes readily with the bath. Later it was found that
the titanium alloys more easily with manganese, but that it is
better to occasion this reaction to take place in the mass of the
molten metal. The resulting improvement is due to the elimi-
nation of oxides and of the nitrogen contained in the metal,
and also, the author believes, to the hydrogen which burns on
Ihe surface of the bath, with its characteristic bluish flame. In
234 ^^^^ Iron and Steel Magazine
practice the addition is made by inclosing the alloy (23 per
cent of titanic oxide, 45 per cent of iron oxides, 25 i)er cent of
ahiminum and 7 per cent of gangue) in a box, attached to a
stem, and placing it in the bath of metal. After a few seconds
the reaction takes place and lasts about a minute, the tempera-
ture of the bath rising and ebullition occurring owing to the
escape of gases, to which the subsequent homogeneity of the
metal is to be ascribed. Numerous investigations were con-
ducted on samples of metal that had undergone this treatment,
the conclusion being drawn that titanium acts principally by
expelling the occluded gases, and eliminating any nitrogen, par-
ticularly in the presence of carbon.
The influence of arsenic on iron and steel was also the sub-
ject of investigation by the author.
PROTECTION OF IRON AND STEEL STRUCTURES*
MEMORANDA OF TESTS
By LOUIS H. BARKER
\ BOUT eleven years ago experimental investigation was begun
'^^ with numerous well-known and established iron paint pre-
servatives, in order to ascertain by actual exposure tests the
best one to resist the destructive action on steel structures of
sulphurous gases in the form of smoke combined with the moist-
ure of steam, and since that time fifty or more paints and com-
binations have been tried. Among them were many kinds of
asphaltum, rubber, graphite, carbon lead and iron paints, and
though the results showed varying degrees of resistance, it is
remarkable that even with three coats of paint not one was found
that did not show rust in less than a year. Of course, it is to
be understood that the exposures were made so as to subject
the test bars to the severest action possible in order to obtain
the quickest results.
In making the first series of tests new steel plates ten inches,
square were used. As, however, the adverse conditions we were
trying to overcome related to rusty steel, which is more difficult
* Read at the June, 1905, meeting of the American Association for
Testing Materials. Abridged.
Protection of Iron and Steel Structures 235
to preserve than new steel, rusty plates were substituted in all
tests thereafter. And to still further endeavor to meet the exist-
ing conditions new plates were hung up and exposed to the smoke
fumes until they became covered with sulphur scale, the
thought being that an oxide scale due to atmospheric exposure
might give different results. This scale or rust formation on
these new plates apparently varied not only in amount, but also
in the time of its formation, supposedly due to different chemical
composition. As this might again give some variations in the
experimental results, in order that all paints should be on as
like footing as possible, angle bars eleven feet long were made
use of, and as before hung in the smoke until rusted, then
cleaned with wire brushes, each foot of the bar painted with a
dift'erent paint and again hung up. The results, however, con-
tinued to be unsatisfactory.
In examinations of the test bars from time to time it was
seen that upon many of them the paint was intact, but with
protruding points which upon being pricked were found to be
small rust formations, pushing up the paint from behind, clearly
indicating that it was not the failure of the paints but the rust
action on the inner surface that caused the damage. As no
rust can form without the presence of moisture, and as all paints
are pervious to moisture (as Dr. Dudley's careful investigations
of the subject have proved) this led to the conclusion that it
would be necessary in some way to tightly seal the surface.
Many kinds of materials for doing this were tried, with as many
-different results, until three years ago it was decided that a
cheap paraffin paper answered the purpose best of all, and since
that time all experimentation has been along that line. The
few test bars that have been brought along and exhibited indicate
the results. Besides the experimental bars referred to, the paper
covering has been tried in a small practical way against smoke
action, and after two years and three months' exposure an exami-
nation of ver\^ recent date shows the outer paint, the paper and
the first or adhesive coat all intact, and in many places where
paper was removed for examination, the adhesive coat not yet
-dry and the surface of steel the same as when painted.
With such satisfactory results from this paper process in
the smoke tests, it was concluded to make a large-scale applica-
tion and severe test on a large number of eye-beams supporting
,^.l
.'■ ^
• . ■ t ^r-^-t '-Vi #
-' - ■ J-. . Vt"
Experimental Bars exposed Eight Months, Lower One-Half Paper,
Upper One-Half Paint, Three Coats
High-Speed Tool Steels 237
a floor over and within a few feet of salt water and upon which
the rust was due not to vSmoke but to the almost continuous
dampness and presence of sewer gases. This was done over a
year ago and up to this time indication of damage of no kind is
apparent.
The mode of application of the paper is as follows: After
the rust is carefully cleaned off by means of stiff wive brushes, a
certain kind of tacky paint is applied, the paper then covered
over and tightly pressed upon the painted surface, the joints of
the paper slightly lapping. As soon as the paper is in place,
it is read}' for the outside coat of paint. It will be observed
that b}' this process, the first coat of paint, the paper and the
coat of paint over the paper can be applied with one scaffold-
ing, thereby greatly reducing the cost, especially in high and
dangerous places.
These experiments, extending over only three 3^ears, are of
too short a duration to determine the value of paper as a protec-
tion for iron and steel, but they certainly bring out the fact, at
least in the case of smoke and gases, that the action begins from
behind the paint and not from in front by the disintegration of
tlie paint.
HIGH-SPEED TOOL STEELS*
By L. P. BRECKENRIDGE, Professor of Mechanical Engineering
Introduction
/^XE of the most striking advances in recent years, from the
point of view of the manufacturer, is the increa,se of the
cutting speeds of machine tools and the consequent marked in-
crease of the output of machine shops. Less than ten years ago
cutting speeds ranged from 5 to 30 feet per minute; now speeds
of 150 feet per minute are frequently employed. The first steps
* An engineering experiment station was established at the Uni-
versity of Illinois by action of the Board of Trustees, December 8, i(;o3.
It is the purpose of the station to carry on investigations along various
lines of engineering, and to study problems of importance to professional
engineers and to manufacturing, railway, constructional, and industrial
interests of the state. The laboratories of the College of Engineering
are being equipped with additional apparatus and facilities to further
such research work.. It is believed that this experimental work will result
238 The Iron and Steel Magazine
in this advance were taken by Messrs. Taylor and White at the
Bethlehem Steel Works during the years 1898-1900, and the
work of high-speed tool steels was first shown to the public at
the Paris exhibition in 1900.
It cannot be doubted that the introduction of high-speed
tool steels is destined to work radical changes in shop processes
and to exert a marked influence on the cost of manufacturing.
Properties of Tool Steels
At the time of Taylor and White's first experiments Mushet
and Jessop tool steels were in general use. These steels were of
the self-hardening type. According to Mr. F. Reiser in an
article on high-speed steel in " Stahl und Eisen," January 15,
1903, thev had the following chemiical composition:
Carbon 2.0% Manganese 2.5% Silicon 13%
Tungsten 5.0% Chromium 0.5%
The self-hardening property is called into play by the man-
ganese, an alloy which favors the combining of the carbon with
the iron. These steels were tempered simply by heating to a
temperature of 1600° F. and then cooling in air. Mushet and
Jessop tools, however, did not prove durable at high speeds,
although they were far in advance of the ordinary carbon steels,
and chromium was substituted for manganese with good results.
The chromium steels required an entirely different treatment,
as was found by Messrs. Taylor and White in their experiments
at the Bethlehem Steel Works.
in contributions of value to engineering science and that the presence
of such investigations will give inspiration to students and add efificiency
to the College of Engineering.
This Circular No. i is designed to furnish the engineers and manu-
facturers of Illinois some information regarding the recent advance in
certain shop processes, — those involving the removal of metal. It will
be followed by bulletins giving a complete account of a series of experi-
ments and investigations now in process in the shops of the mechanical
department.
Credit is due Mr. Henry B. Dirks, fellow in mechanical engineer-
ing, for collecting much of the information contained herein and for
preparing the tables of results.
Copies of this circular may be obtained by addressing the Engineer-
ing Experiment Station, Urbana, 111.
Hi'i^Ji-SpiwI Tool Steels 239
The exact chemical compositions of the new tool steels are
secrets of the separate makers, and probably vary; however,
it is known that the steels contain the following elements in vary-
ing quantities: carbon, tungsten, chromium, manganese, molyb-
denum and titanium. They usually run high in the per cent of
alloy, the Taylor- White steel having as high as 12 per cent of
timgsten and 4 per cent of chromium, while Bohler Bros.'
Styrian steel, according to Mr. Reiser, has a maximum of 28
per cent of alloys. With this increase the carbon element has
greatly decreased; most of it combines with tungsten, chromium
and the other elements at high temperatures, remains in that
state when cooled in an air blast and forms carbides of extreme
hardness and durability at high temperatures. For best results
of toughness and hardness these high-speed steels require for
tempering a temperature of from 2000° to 2250° F., or a white
heat bordering on the fusion point, and are then cooled in an air
blast, lead bath or oil bath, according to the different makers.
Mr. Reiser in his discussion has for this reason correctly nrmied
them " superheated steels."
Advantages of High-Speed Steels
High-speed steels, due to their hardness and durability at
' high temperatures, retain their edge when cutting at extremely
high speeds, cases having been noted in which the tool worked
at dark red heat without losing its edge. As can be seen from
the tables, the speeds obtained are from three to four times
those obtained with ordinary carbon steels. This, of course,
means an increased output for a given shop and a consequent
increase in the returns. This is not the only advantage of high-
speed steel. It has been proved that such steel is more economi-
cal from the power standpoint, a given power removing a greater
quantity of metal per unit of time at high speed than at slow
speed. Of course the total power required is increased, but the
increase is by no means proportional to the increase in the
amount of work done.
There is, however, one condition that must be carefully
considered before the introduction of high-speed steels in a shop.
Machine tools constructed to use the old carbon steels are
limited in capacity and will not stand the heavy stresses to which
240 The Iron and Steel Magazine
they would be subjected if using high-speed steels at maximum
speeds and feeds. This condition, however, is being m.et by
the machine-tool builders, who are now designing and building
especially heavy tools with powerful feed mechanisms with a
view towards obtaining the highest possible efficiency of the
steel used.
Proposed Investigations
While numerous investigations of tool steels have been
made, there is still much to be done along the following lines:
(i) Determination of the most economical speeds at various
feeds and depths of cut for different materials and different sizes
of tools. (2) Determination of the effect of different angles of
rake and clearance on the power required to drive the tool. In
order to obtain some of this information the Mechanical Engi-
neering Department of the University of Illinois has installed an
equiqment for testing high-speed steels. This consists of a
Pratt and Whitney 15 inch by 7 feet 6 inch high-speed lathe
of latest design, driven by a Westinghouse induction motor of
7^ horse-power. The arrangement is such that fifty-six speeds
can be obtained for any given diameter of work. The power
required is measured by a Westinghouse portable polyphase
wattmeter in circuit with the motor, so that readings can be
taken at all times. Arrangements have been made for weighing
all metal removed and for obtaining cutting speeds and feeds.
An apparatus for tempering tools in an air blast has been set up
and the angles of all tools will be accurately measured.
For the present, tests will be made on cast iron of various
grades, a chemical analysis of each of which will be made. In
the future other materials will be tried.
High-Speed Tool Steels
24r
TABLE No. 1.
RESULTS OF EXPERIMENTS
MADE BY
Manchester Municipal School of Technology, England.
Reported by Dr. J. T. Nicolson, Oct. 30, 1903.
Name of
tool steel maker.
Material
operated on.
Size
of cut.
Inches.
Cutting
speed
Ft. per
min.
Material
removed
Lb. per
min.
Dura-
tion
of
trial.
Min.
Samuel Osborn &
Co
Soft C. I.
Axi^ff
84.7
3.175
20
T. Firth & Sons .
i( ((
|xj
53.2
7.33
20
Samuel Buckley .
Medium C. I.
AxA
49.0
1.73
20
t4 ((
(( u
fxi
24.35
3.32
20
C. Cammell & Co.
Hard C. L
r'VxA
31.9
L18
20
({ ((
U ((
m
18.1
2.54
20
Armstrong.
Whitworth & Co.
Soft steel
l\^x\
111.
4.14
20
(1 ((
t( ((
|xi
54.5
7.35
20
C. Cammell & Co.
Medium steel
t\xA
80.
3.17
20
Samuel Buckley
<( i(
m
36.
5.30
20
It n
Hard steel
^x^V
41.2
1.71
20
C Cammell & Co.
(1 ((
fxi
20.
3.00
20
Reported in London Engineenng, Oct. 30, 1903, and in American Ma-
diinifft, Nov, 19 and 26, 1903. The tools in the above cases were all in
nood condition at the end of the trial. The experiments were made on
cast iron and steel cylinders turned in a lathe furnished by Armstrorjg
Whitworth & Co.
242
The Iron and Steel Magazine
TABLE No. 2.
RESULTS OF EXPERIMENTS
MADE BY
BoHLER Bros. & Co., Vienna and Berlin.
Reported by F. Heissig, Jan. 1, 1901.
Name of
tool steel mai<er.
Material
operated on.
Size of
cut.
Indies.
Cutting
speed.
Ft. per
Min.
Material
removed
Lb. per
min.
Dura-
tion
of
trial.
Min.
BohlerBros. &Co.
Cast Iron
JxA-
44.
.735
21
4> i( U
u u
-h^h
43.7
2.59
4
• 4 U (
.( «(
Ux^
44.
1.69
6.5
<. •! % '
.k (k
64X^2^
29.4
1.14
30
1. U i%
kk (•
-i^^h
45.9
3.86
4
.t (k .1
k •
/^XjV
45.9
4.025
4
.t (. >1
kk .(
3 2X1 6
45.9
1.97
28
.t (< u
kk (>
3^XJg
34.8
.882
15
%l il t(
.( a
iVXi^
45.2
.601
22
■ t >l u
Cast Steel
/2X3V
9.85
.203
114
i 4 ( ( 'u
Forged Steel
e'jXi
98.65
.230
7
t (. k .
kl u
3^2X^6
97.7
4.18
5
k( ; tt
(i ik
ejXj j;
157.6
10.46
2
( ( k . ii
k k k .
64-''-8
66.75
2.39
14
>( ti .4
kk ( k
152.6
4 51
4.5
It 11 ii
• i -k
sXb
36.1
6.05
3.5
(. ti k
kk (k
150.
7.06
3
.( (i k<
k k .
64-*'8
53.5
1.29
8.5
Rep(»rted in t^UOd rf- Eim}, Jan. 1, 1901. Bohler Bros. .S: Co. are
represented in the United States by Houghton »& Richards, Boston,
Mass., and their steel is known as Styrian steel.
High-Speed Tool Steels
243
TAHLE No. :i
RESULTS OF DAILY WORK
AT
Union Pacific Railroad Shops, Omaha, Nebraska.
Reported by Henry H. Suplee, July, 1903.
Name of
tool steel
Machine
tool used.
Material
operated on.
Cutting
speed.
Ft. per
min.
Size of
cut.
Inches.
Re-
marks.
All Novo
32-inch
Pond lathe
Piston valve
bushing
Soft cast iron
74
»k k >
(k u
4'' Piston rod
No. 1
Scrap iron . .
18
iXi^,
\i H
27^-incb
Pond lathe
Crank pin
No. 1
Scrap iron . .
26
ixi
l» a
Niles vert,
boring mill
Locomotive
driver tyre
Tyre steel
40
Limit of
belt on
machine
t. n
Bullard vert,
boring mill
Piston head
Cast iron . . .
20
32-^8
it a
Beraent-Miles
hor. cylinder
boring mill
19-inch Cylin-
der. Cast iron
(very hard) . .
18
Limit of
motor
It t 4
88- inch Pond
driver lathe
Driver tyre
hardened by
sliding on sand
24
3v 3
8 -'^3 2
Limit of
machine
(• l(
30-foot
Pond planer
Connecting
rod. No. 1
Scrap iron . .
15
9 yl
1 6-^4
Reported in the Proceedinyfi Institution of Mechanical Engineers,
July, 1903, and in London Engineering, Julv 31, 1903.
244
The Iron and Steel Magazine
TABLE No. 4.
RESULTS OF EXPERIMENTS
MADE HY
Berlin Section, Vereines Deutscher Ingenieure,
Reported in their Pruceedint^s, Sept. 28. 1901
Name of
tool steel.
Material
operated on.
Size of
cut.
Inches.
Cutting
speed.
Ft. per
min.
Material
reniov'd
Lb. per
min
Dura-
tion of
trial
Min
Poldi-Schnell -
dreher. .
Siemens-
Martin Steel .
3 V 1
194
3.33
20
U ('
(; 11
1 1 Y 1
r, 4 ^ I 6
136
4.84
63
Boliler Rapid .
't a
T6*32
126
5.92
95*
(( ((
U .i
B4X8
45.2
3.21
113
Poldi-Diamant .
u u
1 5 V 7
64^64
35.5
2.64
140
Bohler Rapid .
i( u
h^i.
31.6
4.71
120
Boliler
Titan-Boreas .
li (I
mi*
17.4
1.43
120
Bohler Rapid
Cast Iron
9 V 3
72.9
7.39
59
Poldi-Diamant .
u u
16^6 4
50.3
5.21
61
Bohler Rapid . .
i( u
3 3 y 5^
f. 4 ■'^ 3 2
47.3
14.43
120
Bergische-Stahl
Industry
H ((
37.4
4.61
120
Poldi-Diamant
n u
3 3y_»_
6 4 *^ 6 4
35.4
7.39
120
(( it
»( <(
5 Y 9
35.4
5.65
103
Bergische-Stahl
Industry . .
(( u
33 Y 5
6 t^3 2
33.5
10.45
30
u u
(( u
ifxj\
33.5
4.31
148
Poldi-Diamant .
(( <(
I'^f.Xe'?
27.6
5.74
190
Bohler-Rapid
Cast Steel
7 Y 1
47.3
2.45
120
U (i
U ((
7 Y 1
3 5-^1 a
33.5
1.45
120
Poldi-Diamant .
U 41
1 oy "
:i 2 ■'^ 3 2
31.55
2.34
100
Bergische-Stahl
Industry . .
U •(
25y ^
r, 4 ■'^ t> J
26.6
2.15
120
(( ((
U ii
C, 1 ^ f, 4
19.71
2.15
120
Bohler-Rapid . .
U (t
r.Y -^
17.73
2.11
120
Hii^h-Spccd Tool Steels
245
TABLE No. 5.
RESULTS OF TESTS
made before
Representatives of Technical Press at Bethlehem, Pa.,
August, 1900^
Name of tool steel.
Material
operated on.
Size of
cut.
Indies.
Cutting
speed.
Ft. per
min.
Durji-
tion (»f
trial.
Min.
Taylor- White . . .
Mushet
Taylor-White . .
"Mushet
Taylor-White . . .
Mushet
Unusually liard
tool steel ....
Cast i ron ....
Soft machine
steel
15
15
50
50
150
150
15
?
20
If
15
h
Reported in the American Machinist, August 16, 1000.
TABLE No. 6.
Table Showing Increase in Cutting-Speed, Etc., at Bethlehem
Steel Co., since the Introduction of the
Taylor- White Process.
Average.
Oct. 25,
1898
May 11,
1899.
Jan. 15,
1900.
Gain in %
cut of 3rd
over 2nd.
Gain in %
cut of 3rd
over 1st.
Cutting speed
Depth of cut
Feed ...
Lb metal re-
moved per hour
8Mr^
.23^'
.07^'
3L18
21'-9^'
.278^'
.0657^'
181.52
25^-3^'
3(K'
.087'^
137.3
16«
S%
32%
68%;
183%
30%
24%
340%
Average metal removed per hour per tool (round nose)=310 pounds
in April, 1901, and has probably increased some since tnat time.
Reported in American Machinist, August 16, 1900.
246
The Iron and Steel Magazine
TABLE No. 7.
RESULTS OF TRIALS
MADE BY
Messrs. Armstrong, Whitworth & Co., with Twist Drills,
Reporied by Mr. J. Gedhill, Dec. 4, 1903
Name of
tool
steel.
Material
operated
on.
Size of
drill.
Inches.
R. P.
M. of
drill.
Travel of
drill.
In. per
min.
Duration.
.Remarks.
'A. W."
Cast iron
block 4^^
thick . . .
i
525
13i
Drilled
several
holes
Drill
uninjured
t
u
i
360
6
Drilled
137 holes
Not
reground
> I
(t
1
240
4|
Drilled 76
holes
(•
Steel plate
V thick
1
250
5
Drilled
150 holes
Required
grinding
1 1
Steel gun
cradle
y thick
2
80
f
Drilled
124 holes
Reported in Engineering, (London) December 4, 1903.
ABSTRACTS *
{From recent articles of interest to the Iron and Steel Metallurgist)
ALUMINUM-ZINC ALLOYS. E. S. Shepherd. '' Journal of
Physical Chemistry," June, 1905. 4,000 w., illustrated. — Under
the name of Alzine, and Sibley Casting Metal, aluminum-zinc
allovs have been suggested for making castings for which pure
aluminum is unsuited. They are said to expand on freezing,
and to thereby yield sharper castings.
From his experiments, the author infers that this series of
alloys [) resents no so-called definite compounds. There are
two series of solid solutions, that of zinc in aluminum having
a limiting concentration of about fifty per cent zinc, and that of
aluminum in zinc of about four per cent aluminum, at the tem-
perature of 217°. Below this temperature the reaction proceeds
too slowly to permit of accurate determination, and it follows
that the above concentrations are those which are present in
the alloys as met in practice, a,nd he finds his contention sup-
ported both by pyrometric and microscopic data.
The author further says: '^ In so far as we may predict
them, the mechanical properties will not show any great differ-
ence between the cast and annealed alloys. This is due to the
fact that the alloys apparently reach equilibrium readily, and
any but very rapid chilling will allow the metal to come to
equilibrium. Annealing increases the size of the crystals very
greatly indeed and renders them more brittle. It is a fact
well known to technical men, that these alloys become crystal -
* Note. The publishers will endeavor to supply upon request the full
text of the articles here abstracted, together with all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60
cents, "D" 80 cents, "E" $1.00, "F" $1.20, "G" $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cases
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
ing upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.
247
248 The Iron and Steel Magazine
line under repeated shock, and break. It is for this reason
that they have been largely displaced by the Al-Sn alloys, and
are now used chiefly for ornamental castings, and for meter
cases." No. 406. C.
Chain Making by Electric Welding. Andris-Jochams.
"The Iron Age," July i-^, 1905. 4,400 w., illustrated. — The
author describes the application of electric welding to the manu-
facture of chains, a method patented by E. F. Giraud, of France.
Claims for the Giraud process are economy, rapidity and superi-
ority of production.
The author concludes as follows: "The chain-making in-
dustry in the United vStates represents an immense investment
in improved machinery and processes. The average annual
production of welded chain in this country is at least 50,000
tons. The adoption of the electric welding process would mean
the dismantling of over 20 large modern chain shops and might
result in the concentration of the production into fewer shops."
No. 407. B.
The Growth of Large Gas Engines on the Continent. Ro-
dolphe E. Mathot. A paper read before the Institution of
Mechanical Engineers. 14,000 w., illustrated. — The author
describes the principal types of large gas engines constructed
on the Continent and the methods and principles of the makers.
No. 408.
Cooling during Quenching of Steel. P. Lejeune. " Rev.
de Metallurgie," April, 1905. 5,000 w., illustrated. — The author
studied the cooling of steel during quenching by means of ap-
paratus due to Le Chatelier, which acts somev/hat similarly to
that of Saladin. He finds that quenching in a small bulk of
mercury, e. g., as practiced for surgical instruments, is slower
than in water. The maximum rate of cooling from the quench-
ing temperature (about 850° C.) is indicated between 500° and
600°, but occurs at a much earlier stage in water and mercury
quenching than with some other liquids; the thermo-couple,
however, is embedded in the specimen. Using mixtures of
water and glycerine, it was found that the duration of quenching
varied in a somewhat similar way to the viscosity. The follow-
Abstracts 249
ini^ figures show this variation, the volume of Hquid in each
case is 2 Hters: glycerine, per cent, In^ weight, 2, 10, 20, 50; rela-
tive viscositv (water = i), 1.05, 1.13, 1.33, 1.6. The temperature
fell from 700° to 100° in 6.5, 8.2, 10.3, 11.3 seconds respectively.
The specific heat of the liquid affects the duration of cooling.
Alcohol and xylol were used, and a comparison was also made
'oetween a bath of oil and one of equal viscosity consisting of
90 per cent glvcerine in water; 27.5 and 1 1.3 seconds respectively
were required for the fall of temperature from 700° to 150°.
The shape of the cooling curves is affected by ebullition of the
liquid and by the occurrence of a jacket of vapor about the
specimen. The latter were in the form of cylinders of equal
diameter, and the duration of quenching w^ould seem to vary less
rapidly than the length and more rapidly than the ratio of
volume to surface. — vScience Abstracts, July 26. No. 409. C.
Iron-Ore Deposits in Foreign Countries. '^ The Engineer,"
June 23, T905. 1 ,800 w. — Analysis of a report of nearly 300 pages
compiled at the board of trade (England) from a very large
number of diplomatic and consular reports. No. 410. B.
16,000 Horse-Power Vertical Rolling-Mill Engines. " En-
gineering," June 23, 1905. 2,100 w., illustrated. — The article
describes a powerful set of three-cylinder vertical reversing
engines constructed for the Britannia Works of Messrs. Dorman,
Long & Co., \)y Davy Brothers of vShefheld. They are be-
lieved to be the most powerful engines that have ^■et been
built for the purpose. No. 411. B.
The Relative Economy of Blast-Furnace Gas Engines and
Steam Engines in the Lorraine District. Dr. Ehrhardt. " Stahl
imd Eisen," June i, 1905. 3,000 w. No. 412. D.
METALLURGICAL NOTES AND COMMENTS
Few engineers are more widely known
Rossiter^Worthington ^^^^ ^^ ^ ^ Raymond, the esteemed
secretary of the American Institute of
Mining Engineers, and the features of his photograph reproduced
on our frontispiece will undoubtedly be familiar to most of our
readers. The short outline we give below of his professional
career testify to a useful, successful and therefore inspiring life.
Through his high professional accomplishments and his attrac-
tive and gifted personality, Dr. Raymond has attained marked
distinction as geologist and mining engineer, as writer and editor,
as lecturer and presiding officer and more especially as secretary
of one of the largest and most important engineering societies in
the world, a position for which he possesses, in so remarkable
a degree, the desirable qualifications.
Rossiter Worthington Raymond, Ph.D., was born April 27,
1840, at Cincinnati, Ohio. In 1852 he graduated from the Poly-
technic Institute, Brooklyn, N. Y., and from 1858 to 1861 studied
at the University of Heidelberg and Munich and at the Mining
Academy of Freiberg, in Germany. From 1861 to 1864 he
served as officer in the United States Army, and from 1864 to
1868 practiced in New York City as a consulting mining engineer
and metallurgist. From 1868 to 1876 Dr. Raymond occupied
the position of United States Commissioner of Mining Statistics
for the states and territories in and west of the Rocky Mountains,
his annual reports having been published by the government
in eight octavo volumes. From 1870 to 1881 he was professor
of economic geology at Lafayette College, Easton, Pa. Dr.
Raymond was an original member of the American Institute of
Mining Engineers, founded in 1871, and was that year elected
vice-president of that society, while from 1872 to 1875 ^^^ served
as president; from 1876 to 1877 again as vice-president; and,
since 1884, by annual election he has been the secretary of the
institute, editing the yearly volumes of " Transactions."
250
Mctallitrf^ical Xotcs and Comments 251
From 1867 to 1890 Dr. Raymond was editor of " The Engin-
eering and Mining Journal " of New York (called for the first
two years of that period " The American Journal of Mining ").
In 1873 he was United States commissioner to the Vienna Inter-
national Exposition. From 1875 "to 1895 he was consulting
engineer of the firm of Cooper & Hewitt, iron and steel manu-
facturers, and for the greater part of that time acted as an
assistant of Hon. Abram S. Hewitt in the Cooper Union, of
which he directed the free popular scientific evening lectures
until it was assumed by the board of education. From 1885
to 1889 he was one of the New York State electric subway
commissioners for the city of Brooklyn, whose final report,
written by him, was widely republished as the best statement of
the problem of municipal electrical engineering as it then existed.
Dr. Raymond is a member of the New York State and of the
federal bar, practicing chiefly in the departments of mining
and patent law. His essays on mining law are quoted as au-
thority. In 1904 he was lecturer on mining law in the Law
School and Mining School of Columbia University.
He is an honorary member of the Society of Civil Engineers
of France, the Iron and Steel Institute, and the Institution of
Mining and Metallurgy of Great Britain, the American Philo-
sophical Society and other scientific bodies.
Outside of his professional work, Dr. Raymond has published
several volumes of stories, notes of travel, criticism, etc., and
numerous biographical sketches of distinguished men personally
known to him, including Alexander L. HoUey, Peter Cooper,
Abram S. Hewitt, Eckley B. Coxe and many European technical
authors and teachers.
The autumn meeting of the Iron and Steel Insti-
Iron and Steel ^^^^ ^^n ^^ ^^^^ ^t Sheffield on September 26,
Institute rT>i r 1, • 1
27 and 28, 1905. The followmg papers have
been offered for reading:
T. " The Metallurgical Department of Sheffield University,"
by Prof. J. O. Arnold (Shefheld).
2. " The Thermal Transformation of Carbon Steels," by
Prof. J. O. Arnold and A. McWilliam (Sheffield).
3. " The Nature of Troostite," by Dr. Carl Benedicks
(Upsala).
252 The Iron and Steel Magazine
4. " The Occurrence of Copper, Cobalt and Nickel in Amer-
ican Pig Irons," by Prof. E. D. Campbell (Ann Arbor, Mich.).
5. " Pipe in Steel Ingots," by J. E. Fletcher (Sheffield).
6. " Steel for Motor-Car Construction," by L. Guillet (Paris).
7. " The Presence of Greenish-Colored Markings in the
Fractured Surface of Test Pieces," by Capt. H. G. Howorth,
R.A. (Sheffield).
8. " Over- Heated Steel," by Arthur W. Richards (Grange-
town) and J. E. Stead, F.R.S. (member of Council).
9. " Segregation in Steel Ingots," by B. Talbot (Middles-
brough) .
10. " A Manipulator for Steel Bars," by Douglas Upton
(J arrow) .
11. " Machinery for Breaking Pig Iron," by Cecil Walton
(Whitehaven).
12. " The Influence of Carbon on Nickel and Iron," by
George B. Waterhouse (New York).
The following papers were read before
The Mining and ^■\^^ metallurgical section of the Congress
Metallurgical Congress ,, 11^ .tw t-.i-
^.v recently held at Liege, Belgium, m con-
nection with the International Exposi-
tion now being held in that city.
'' The Use of Poor Caking Coals for Coke-Making," by M.
Hennebutte.
'' The Cleaning of Blast-Furnace Gas," by M. Bain.
'^ The Influence of Titanium on Pig Iron and Steel," by M.
Delville.
" The Manufacture of Cement from Blast-Furnace Slag,"
by the Chevalier Cecil de vSchwarz.
'' The Employment of Blast-Furnace Slag for Portland-
Cement Manufacture," by Dr. H. Wedding.
'' Review of the New Methods of Making vSteel in Open-
Hearth Furnaces," by P. Ackers.
'' Special Steels," by M. Guillet.
'^ Methods of Preventing Piping in Steel Ingots," by M.
Daelen.
'' Effect of Liquid-Air Temperatures on the Properties of
Iron and Its Alloys," by Mr. R. A. Hadfield.
Metallurgical Notes and Comments 253
" The Sheariui^ and Cuttin": of Metals by Oxygen," by
M. jottrand.
" Double Tempering for Large Steel Forgings."
" The Microscope in Metallurgy," ])y M. Le Chateher.
*• The Electric Smelting of Steel," by M. Gin.
" The Electric Furnace in Metallurgy," by M. Pitaval.
" The Cermak-Spirek Furnace for Roasting and Calcin-
ing Minerals," by M. Spirek.
Two of these papers will be found reproduced in the present
issue of The Iron and Steel Magazine, while the others will be
printed in subsequent numbers..
The Chemist and His Work. — One of the most important
assistants to the metallurgist is, without doubt, the chemist,
who ma\' well be called his aide-de-camp. The results received
from him are now taken so much as a matter of course that it is
almost impossible to believe that but little more than twenty
years ago the laboratory, if it existed at all, was surrounded
with mystery. A " full " analysis of a specimen of steel
was then spoken of with bated breath, and its accuracy w^as
doubtful.
When one peruses a work on metallurgy of even so late a
date as a hundred years ago, the extraordinary designations and
terms employed seem to indicate that metallurgy was regarded
as a branch of cabalistic art. Certainly before the year 1750.
or about that date, the art of metallurgy outside the production
of the simplest products seems almost to have been classed
with witchcraft and sorcery.
Past metallurgical literature shows how impossible it was
for the workers in those days to advance until the alchemist
passed aw^ay and the chemist appeared on the scene. It has,
therefore, been largely due to the chemist that the Egyptian
darkness prevailing so long in the world's history has been
cleared away, thus making the first real progress in our era
possible. All honor to the noble little band of chemists who
led the way to that better state of things from which metallurgy
in due time received its share of the progressive benefits re-
sulting from their work I They came largely from Sweden
towards the middle and the end of the eighteenth century,
2 54 The Iron and Steel Magazine
whilst later, those of Germany, France and England joined in
the good work.
A modern metallurgical chemist has to take a wide range
of view. He nuist not only be well gifted in the art of the
laV)orator3^ at large, but must keep an intelligent eye upon the
many ramifications of his special branch. After showing that
the raw materials which pass under his notice are satisfactor}%
he may for a time lose touch with the actual manufacturing
processes, but the final product mxust come under his hands
again. Unlike the mechanical engineer, whose work is more
or less finished with the completion of his contract, it is to
the chemist the metallurgist has often to turn, long after the
product has been in use, to show whether his work has resulted
in failure or success.
It is, therefore, a matter of no little anxiety for the chemist
to certify that the products placed before him are satisfactory ;
for great as have been the advances in laboratory analytical
methods, we are yearly finding that as products become more
complex so must our laboratory respond to the call made upon
it and rise to meet conditions involving difficulties of no mean
order.
The laboratorv is often considered to be a mere recording
instnmient; but it is far more than that. There are problems
in steel manufacture awaiting solution v^hich can only be
solved by the chemist; but his methods, although they are
aided by microscopy, are still far from perfect. Take one
example alone, — that of the correct determination of the various
forms of carbon in steel. For many purposes we want far
more than a mere laboratory report of the percentage of car-
bon present in a specimen — that is much less than half the
knowledge we need. We must know in what forms the carbon
exists. Abel, Arnold,. Blair, Brinell, Campbell, Campredon,
Jiiptner, Ledebur, Edward, Riley and Stead have done great
work in this direction. The complex combinations now met
with in alloys or special steels will be appreciated when it
is remembered that there are often more than ten different
and important elements in a single alloy combination. This
will give some idea of the difficulties to be encountered. It
is thus found that our present methods are not too reliable,
and there is room for much more light in this direction.
Metallurgical Notes and Comments 255
The subject is of such importance that if a special com-
mittee were formed to take up this matter it might do great
service to the world at large whilst benefiting the metallurgist
more directly. vSuch information would enable us to secure,
unerringly, important practical results that are now obtained
in far too haphazard a manner. We have at present no
means of properly determining these com^binations by the
method which should be the simplest, best and most trust-
worthy,— that is, chemical analysis. Mysterious failures still
occur, and one of the helps to solve this problem will be to
know the full why and wherefore of the exact composition
of our steel. It is for these reasons, amongst others, that
I place the responsibility of the mietallurgical chemist so
high. When once he recognizes what is needed from him, he
will, we mav be sure, rise to meet the want, as he has done
in the past. Since the great work of Abel and Ledebur there
has been too great a lull in the demands we have made upon the
chemist. We have been content to accept his results as a mere
laboratory rettirn of the percentages of certain elements, often
not recognizing that the various complex combinations of car-
bon are of vital importance.
Excellent work is now ]:)eing undertaken conjointly by Dr.
R. T. Glazebrook, F.R.S., of the National Physical Laboratory,
and Mr. H. Le Chatelier in their researches '^ On the Constituents
of Steel." The first object of this investigation is to find the
conditions under which quenching for hardening steel should
be carried out, with a view to determine the nature and
character of the various micro-structures of steel as ordinarily
accepted. Although these at present are purely theoretical,
no doubt disciission will render excellent service in enabling
investigators to understand the different structures of steel
obtained by heat treatment.
In America a movement is in progress for improving and
sy stigmatizing chemical analysis. It is proposed that the
National Bureau of Standards at W^ashington should take
up this work. An influential committee has been formed,
including, amongst others, Dr. C. B. Dudley. The following
procedure has been recommended:
(i) To impress on chemists the necessity, by using new
methods and ymre materials, of arriving at greater uniformity.
256 The Iron and Steel Magazine
(2) Tt) try to discover to what the present lack of uni-
formity is due.
(3) To test, in cooperation with the National Bureau of
Standards, various methods to see if they are suitable for gen-
eral use.
(4) To prepare samples of different character, the composi-
tion of which is known exactly.
(5) To distribute such samples to persons wishing to test
the accuracy of their methods of analysis. From R. A. Hadfield's
Presidential Address, Iron and Steel Institute, May, 1905.
The Knoth Slag Process for Manufacturing Steel. — On
May 2, 1905, letters patent were issued to Henry Knoth, for-
merly superintendent of the steel plant of the Tennessee Coal,
Iron and Railroad Company, at Ensley, Ala., and now superin-
tendent of the Monterey Steel Plant, at Monterey, Mexico, for
a process which has proved successful in increasing the output
of steel while reducing the cost of production. Mr. Knoth,
who has given considerable study to the manufacture of basic
open-hearth steel, reached the conclusion that the present
method of manufacturing steel could be greatly improved if
a liciuid basic slag could be provided at a minimum expense
to start the heat quickly into action. When the molten slag
is tapped from an open-hearth steel furnace and thrown away
there is a loss not only of considerable basic materials but also
of the heat in the slag. OlDviously, if this heat and these basic
properties can be successfuly utilized in purifying other heats
an economy in manufacture will result. According to the Knoth
slag process the liquid slag resulting from an initial heat, pre-
pared in the usual manner in a basic open-hearth furnace, is
used continually to purify subsequent heats by being returned
to the same furnace, unless it is stopped for considerable
repairs, and in that event to any other furnace then ready for
it. The losses in the basic properties of the slag by continually
purifying heats are replaced by lime or other desirable fluxing
materials.
Preferably the unpurified metal to be acted upon by the
molten slag is introduced into the furnace in a molten condition,
or, better, is first blown in an acid converter. In both cases
the reactions will at once set in in the bath, since the slag is in a
readv condition. The duration of the heat is thereby considerablv
, Meiallnrgical Notes and Comments 257
reduced. This process is being used at the Monterey Steel Plant,
where it has uniformly given excellent results. Thus, 24 tons
of pig iron were melted with 6 tons of scrap without additions
and tapped with i per cent carbon into an open-hearth furnace.
The slag from a previous heat in this furnace was returned
thereto in liquid condition with about 1,000 kg. of limestone.
The liquid slag inmiediately acted on the limestone, and after
the addition of some fluorspar the reaction began at once.
After adding ore several times the heat was tapped in two
hours and thirty-five minutes. This process was repeated a
number of times with substantially similar results. Below
are the analyses of the unpurified metal, the slag and the steel
as tapped in three succeeding heats, the times for the heat
being given.
Dlratiox of the Heat from Charging of the Liquid Material till
Tapping
Xo. I No. 2 No. 3
2 hours 35 minutes 2 hours 40 minutes 2 hours 5 minutes
Analysis of Unpurified Metal Charged
No. I No. 2 No. 3
Per cent Per cent Per cent
Silicon Trace Trace Trace
Manganese 0.14 0.16 0.15
Phosphorus 0.115 0.121 0130
Sulphur 0.061 0.065 0.064
Carbon 1.05 0.96 0.89
Analysis of the Slag
No. I No. 2 No. 3
Per cent Per cent Per cent
SiO, 12.96 12.77 11.76
AI2O3 7.89 8.1 1 7.93
Fe 15.40 11.48 14-39
CaO 48.47 47-50 44-79
MgO 6.62 7.28 8.51
MnO 2.00 2.38 2.95
S 0.38 0.45 0.46
P2O5 1.80 2.96 4.35
Analysis of the Steel
No. I No. 2 No. 3
Per cent Per Cent Per cent
Manganese ,, . .0.38 0.32 0.45
Phosphorus o.oi o 006 0.012
Sulphur 0.045 0.038 0.049
Carbon o.io 0.09 0.23
258 The Ir Oft 'and Steel Magazine ,
These analyses are furnished by Franz Putsch, chief chemist
of the Monterey Steel Company.
The process will operate most successfully when the pig
metal is treated in an acid converter, and it is clear from
the results above obtained that where the unpurified metal is
blown in the converter to i per cent carbon and then charged
in the open-hearth furnace and treated in accordance with the
Knoth slag process a production of 200 tons of steel within
twenty-four hours can be easily obtained from a 30-ton furnace.
The Knoth process will be of considerable importance to
Southern steel manufacturers. It is believed that the saving
m basic materials largely offsets the cost of blowing the metal
in a converter and would cheapen the duplex process, such as is
now being successfully carried out at the Ensley steel plant.
In addition to this the other advantages of the process consist
in the increase of production, the short time of the heats in the
furnace and a correspondingly increased life of the furnace
hearth, the opportunity to repair the furnace bottom between
heats without interrupting the continuity of the process, the
utilization of all the basic properties in the slag and the reduc-
tion of the quantity of material thrown on the slag pile and han-
dled at an expense. ^^ The Iron Age," July 13, 1905.
The Iron Pillar at Delhi. — When one goes to India he is
apt to have his ideas of modern achievement and civilization
considerably changed. Americans usually think they are the
only ones who know how to produce large forgings, and yet on
the plains of Delhi, the ancient capital of India, near the Kutub
Minar, stands an immense iron pillar which is the most curi-
ous antiquity of all India. It is a wonderfully fine forging, 50
feet long and 16 inches in diameter. It rises 22 feet above
ground — the remainder is below ground. There are several
weird legends connected with its history, and its exact date is
unknown, but it is certain that it extends back into the past to
a date before our boasted Western civilization even had a begin-
ning. It records its own history in Sanskirt in a deeply cut
inscription on its face, and students assign the third or fourth
century after Christ as the date of this inscription. The Colossus
of Rhodes and the statues of Buddha described by Iliown
Thsang, the old Chinese traveler, were of brass or copper, hollow.
."l/t7i7////;\'/V(7/ A'Otcs mid Comments
259
IS a
and of pieces riveted together, but this " PiUar of Fame
soHd shaft of wrought iron.
In one of the temples of India there are some iron beams
supporting a portion of the building which are almost as re-
markable forgings as the pillar itself, and probably as old. These
Iron Pillar of Delhi, with Arch of Kutub to the Right
beams are about twice as deep in the center as at the ends,
showing that the people of that remote age understood how to
distribute the material in the beams to the best advantage. It
is exceedingly interesting to know that these great forgings
which were made over fifteen hundred years ago still stand as
a monument to the ancient industries of India.
m-
O
<
CO
o
to
?5
O
tS3
U.
o
a:
I
X
O
e
<
CO
fd
<
pi
M
%
K
O
2 62 The Iron and Steel Magazine
India boasts of exceedingly rich deposits of iron, but un-
fortunately the iron ore and flux are widely separated and
the coal is so high in ash that it has not been possible to work
deposits by modern methods. The most primitive furnaces
w^ere probably short shaft, built in the underwind side of a
hill, the wind furnishing the blast. Charcoal was used as
fuel and the ores reduced in small quantities. The thick-
masses of iron were raked out of the bottom of the furnace
and forged into shape so that they could subsequently be
welded into large m.asses if required. Later the furnaces w^ere
equipped with bellows, and their height greatly increased until
it reached about lo feet, and the output was greater than that
of any other primitive iron-making furnace. It is interest-
ing to know that cast iron was probably first made in these
furnaces of ancient India, but it was made as the result of an
accident and was always thrown awa}^ with the slag. When
the draught became especially good and the furnace was very
hot it would sometimes make a few htmdred pounds of cast
iron, which would run out of the slag and be thrown away.
These tall furnaces are still used in some remote districts of
the Him.alayas, but thev are fast being displaced by iron and
steel products imported from Europe and America.
The iron pillar at Delhi has always been bright, and some
writers attribute this to the fact that certain of the ])ilgrims
divined this pillar as a part of their religious observances, and
claim that the grease and oil applied to it in this wa}^ pro-
tected it; but it is probably true that the surface is covered
with a black oxide rust Avhich has protected it from further
oxidation. Several attempts have been made to obtain a sam-
ple of the iron for analysis, but as the natives regard the pil-
lar with superstitious awe, no complete analysis of the material
has been made.
Wondrous does this tale of a past civilization become when
one remembers that it has been only in late years that any
modern nation has been able to create anything like so massive
a pillar of wrought iron. '^ Iron Trade Review," July 6, 1905.
Blast-Furnace Practice. — The practical application of the
pyrometer to the determination of the temperatures of the
zones of chemical changes which are found to exist in the blast
MctaUitrgical Xotcs and Comments 263
furnace has enabled the author to prepare the accompanying
diagram (]>ages 260 and 261).
The temperature eurve shows the gradual increase in tem-
perature of the materials of the charge as they descend in the
furnace through the different zones. The chemical changes
are clearly indicated by the portion of the diagram marked in
percentages. Increasing or decreasing the height of the furnace
would not greatly modify the diagram, while changes in the
diameter would affect the time during which the materials
take to pass through the zones. It will be noticed that the
temperature line in the neighborhood of the boshes very nearly
coincides with the angle of the brickwork, i. e., about 70°.
In regard to the formation of slag, it will be obvious that
this can onlv proceed as the combining materials intermingle
in close contact, and reach a part of the furnace where the tem-
perature is sufficiently high. A reference to the diagram will
explain the reason why the quality of the iron is affected by
allowing the stock to fall too low in the furnace; the tempera-
tures in the heat interception and fusion zones would remain
constant, but the volumes of the zones of preparation and re-
duction would be considerably curtailed. Horace Allen, in
" Iron and Coal Trades Review," July 7, 1905.
Percentage of Iron and Steel and Other Leading Products
of the United States Steel Corporation and Independent Producers
in 1902, 1903 and 1904. — In the following table we compile
from our Annual Report for 1904 the percentage of the total
shipments of iron ore from the Lake Superior region by the
United States Steel Corporation in the calendar years 1902, 1903
and 1904, as compared with the shipments of iron ore from the
same region by all other companies, firms and individuals; also
the percentage of the total production of iron ore, coke, pig iron,
steel ingots and castings, all kinds of finished rolled iron and steel
and wire nails by the corporation and by all independent pro-
ducers for the same vears.
264
The Iron and Steel Magazine
The percentage of shipments and
I
902
1903
1904
production b> the L iiited Mates
Steel Corporation and by inde-
pendent producers.
u. s.s.
Corp.
Independ-
ents
u. s.s
Corp.
Independ-
ents
u, s. s.
Corp.
Indef>end-
ents
Shipments of iron ore from
Lake Superiur ....
Total production of iron ore,
Production of coke . . .
604
45.1
37-4
39.6
54-9
62.6
58.8
43.8
34.2
41.2
56.2
65.8
53.8
38.0
36.6
46.2
62.0
634
Ijessemer, basic, forge, foun-
dry and all other pig iri^n,
Spiegeleisen, ferro-manga-
nese.and ferro-phosphorus
44.3
81.0
55-7
19.0
39-9
81.0
60.1
19.0
44-3
70.5
55-7
29.5
Total pig iron, including
spiegeleisen, etc. . . .
44.7
55.3
40.4
59.6
44.6
55-4
Bessemer steel ingots and
castings
Open-hearth steel ingutsand
castings
73-9
52.4
26.1
47.6
72.0
51.0
28.0
49.0
69.0
50.4
31.0
49.6
Total steel ingots and
castings
65.7
34.3
63.5
36.5
61.0
390
Bessemt r steel rails . . . | 65.4
Structural shapes . . . . ! 57.9
Plates and sheets, excluding }
nail plate I 59.4
Wire rods | 71.5
Bars, skelp, nail plate, open-
hearth and iron rails, etc. { 3 1. 1
34-6
42.1
65.6
60.3
34-4
39.7
1
57-2
55-1
40.6
28.5
59.9
73.1
40.1
26.9
580
7.3
68.9
29.8
70.2
28.6
42.8
44-9
42.0
28.7
71.4
Total of all finished rolled
products
50.8
49.2
51 2 48.8
47.8
52.2
Wire nnils
648
35.2 70.6
29.4 67.0 I 33.0
In the three years covered by the table the percentage of
the production of iron ore by the United vStates Steel Corpora-
tion fell from 45.1 per cent in 1902 to 43.8 per cent in 1903 and
to ;^S per cent in 1904. The decrease in 1904 was due principally
to the large stock of iron ore which had been carried over from
Mttallitrgical Xotcs ajid Coftwioits 265
190^^ So also was the decrease in the shipments of iron ore in
1904. The corporation's percentage of the production of pig
irt)n was about the same in 1904 as in 1902. The corporation
was a large buyer of pig iron in the three years under review, its
furnace capacity not being equal to its converting and finishing
capacitv. In steel ingots and castings a decrease of 4.7 per
cent is shown in 1904 as compared with 1902. Bessemer steel
rails show a decline from 65.4 per cent in 1902 to 57.2 per cent
in 1904, the decrease in the latter year being caused mainly bv
the competition of the Lackawanna Steel Company (independ-
ent), which was an active competitor for Bessemer rails in 1904,
which it could not be in 1902 and during the greater part of
1903. while building its new plant. The increased competition of
open-hearth steel rails in the South was also a factor. All kinds
of finished rolled iron and steel, including rails, show a decrease
of 3 per cent in the period covered by the table, although wire
nails show an increase of 2.2 per cent in 1904 as compared w4th
1902. In structural steel the corporation lost nearly three points
in 1904 as compared w^ith 1902. In wire rods and plates and
sheets there were slight losses by the corporation in 1904 as
compared with 1902. ^' The Bulletin," American Iron and
Steel Association, July 15, 1905.
Blast-Fumace Filling Apparatus. — John W. Seaver. assignor
to Wellman-Seaver-Morgan Company, Cleveland, Ohio, has
patented (U. S. 792,735) the method of charging furnaces here
outlined, which is valuable principally because of its flexibilitv,
and because the ore may be run in at any convenient level below
or above the furnace top. A combination of two traveling covers
is placed over the furnace, the edges of the covers extending
down into troughs to. form a water seal, and a traveling lorrv
being arranged to run over the auxiliary cover, which is normally
out of use. Among the several ways in which the apparatus
may be employed in a plant is the one shown in the drawing.
Here the ore is run in on the lorry (17) on the ground level, and
then hoisted (by any convenient means, not shown) to a position
on the traveling crane at the left of the drawing. At the point in
the illustration the lorr\' (17), shown now in side view, is moved
by its motor (18) along the track ( 1 5 ) , so that its discharge nozzle
(21) projects over the endless conveyor (13). The motor (12)
266
The Iron and Steel Magazine
moves auxiliary cover (9), pushing the cover (5) in advance, the
depending edges being then raised out of the water-seal trough.
As the covers move, the conveyor apron (13) conveys the charge
from the lorrv to the end of the auxiliary cover and dumps it
between the covers into the furnace (i). The covers may be
moved back and forth at any convienent speed to distribute the
charge as desired. By extending the lorry's track, several
kinds of ore mav be received from different points and separately
weighed by passing over scale-beams. If it is desired to deposit
the coke on the edges of the furnace the lorry can be provided
with side partitions for the coke, the ore being carried in the
center. '' Engineering and Mining Journal," July 22, 1905.
The Invention of Grooved Rolls. — The "Iron and Coal
Trades Review " publishes the following letter:
'' Whilst abroad in March I observed a statement in the
' Times ' that a monument of Henry Cort has been raised, with
an inscription which added to his well-known claim to be the
inventor of puddling that of being the inventor of grooved
rolls. The inclosed abstract from a very interesting paper,
entitled ' How to Beat the Dutch without Fighting,' by Sir
William H. Bailey, of Manchester, on two volumes by Andrew
Yarranton, written in the year 1676, entitled ' England's Im-
provement by Sea and Land. To out-do the Dutch without
Fighting; to Pay Debts without Moneys,' reads as follows.
Mctallnriiica! A>otcs ajid Cofnmcuis 267
" * He (Yarranton) says: '^ I have noticed the way in which
rings and bolts are made for ships, and we ought to imitate in
this country what they do in Germany, for I have seen bars of
iron made perfectly round, about 12 feet in length and i inch
diameter, and they are sent down the river to Holland, where
they are employed for making bolts and rings. They are
unifonn in size, and are much better for use in shipping than
those which are made with the hammer." It appears these
were made with an engine truly made for the purpose, and
he suggested that such a machine should be introduced into
this country. That is a common rolling mill, and it shows
how far we were behind in iron manufacture in his time when
that was an unknown tool in this country.'
'' It would be interesting to know if any of your readers can
throw further light on this subject, as it is clear that bars 12
feet long and i inch diameter, perfectly round, could not have
been made or rolled other than in grooved rolls.
'' Fred Mills.
'' Ebbw Vale, May 22, 1905."
James M. Swank in the '' Bulletin " of the American Iron
and Steel Association comments on the letter as follows:
"' In Chapter 8 of ' Iron in All Ages ' will be found much
interesting information concerning the use of the rolling mill
^rior to Cort's time. Cort's great invention was the puddling
furnace, not the rolling mill, but he is entitled to the credit of
having improved and enlarged the uses of the latter." ^' Iron
Trade Review," July 6, 1905.
Electric Furnace for Research Work at High Temperatures. —
Dr. Harker has invented a form of electric furnace in which
small samples of metal can be heated to temperatures of 2200°
C. without coming into contact with carbon or any noxious gases.
The conductor conveying the electric current is a tube of similar
composition as the filament of a Nemst lamp. For many pur-
poses the usefulness and life of a furnace constructed in this way
may he. much increased by adopting a '' cascade " system of
heating; that is, the energy supplied may be divided, so that
only sufficient is put through the tubular conductor to raise its
temperature, say, 1000° C. above its surroundings, the sur-
268
The Iron and Steel Magazine
roundings being maintained at 1000° C, thus enabling a tem-
perature of 2000° C. to be attained in the tube without straining:
it unduly. The regulation of temperature in small furnaces of
this type is so perfectly under control that very well defined
melting-points may be taken with very small quantities of sub-
stance. The thermo-electric method has been used in these
furnaces for determining the melting-point of platinum, the
mean result of the experiments giving 1710° C. ± 5° C. ''Tech-
nics," July, 1905.
William Jessop
We regret to announce the death of Mr. William
Jessop, of Thornsett Lodge, Bradfield, which
took place at his residence, after a very long illness, on Tuesdav.
The deceased gentleman was well known as the chairman of
William Jessop & Sons, Lim-
ited, steel manufacturers, of
Brightside, Sheffield. Mr.
Jessop, who was only forty-
eight years of age, was a son
of the late Mr. Thomas Jessop,
who was largely responsible
for the building up of the
great business at Sheffield.
It was founded by a pre-
vious generation of the family
but made its most rapid
development during the time
it was controlled by Mr.
Thomas Jessop, who was a
man of great resource and public spirit. He was master cutler
in 1863, and mayor of Sheffield in 1 863-1 864. Mr. William
Jessop was educated at the Collegiate School, Sheffield, and after-
wards at Repton. From Repton he went to Germany, and
eventually completed his education at Cambridge University.
He did not, during the greater part of his Ufe, take a very active
part in the business at Brightside, although nominally he. was
concerned with the firm, and from 1880 had occupied the posi-
tion of a director. On the death of his father, in 1887, he became
chairman of the company, and remained so until the time of his
death. In addition to his directorship of William Jessop & Sons,
ATctalliiriiical Xotcs cmd Comments 269
lie was also for a considerable period a director of the Sheffield
and Rotherham Bank, and a director of the Yorkshire Engine
Company. Mr. Jessop's only son, Mr. Thomas Jessop, entered
the business so recently as a month ago, after completing his
education. It is interesting to note that the business of William
Jesscp & Sons has existed under its present title since 1830, but
the first William Jessop was in the trade long before then, and
there are records of Jessops making steel in Sheffield as far back
as 1774. Originally carried on in Blast Lane, the business
developed remarkably at Brightside, under the direction of the
brothers, Sidney and Thomas Jessop. The business was con-
verted into a limited company in 1875, and has constantly
grow^n, one of its latest developments being the opening of a
branch establishment in America in 1902. '' Iron and Coal
Trades Review," July 7, 1905.
The Recent Movement of the World's Finished Iron Indus-
try. — It is more than twenty -five years since numerous prophe-
cies were made by '^ men of light and leading " in the iron trade,
which, if they had come true, would have closed all the old-
st\'le mills and forges, not in this country only, but throughout
the world. The sanguine supporters of the new era, ushered in
by Bessemer and Siemens, believed that by this time puddled
iron would be an extinct industry, and we were periodically
regaled with the statement that the labor of the puddler — '' the
hardest work voluntarily undertaken by man," as it was deemed
the correct thing to describe it — was about to end. Never was
the danger of prophecy more signally demonstrated. There has
not only been no extinction of the finished iron industry, but
that industry still yields between five and six million tons a
year of products that are as much appreciated as they ever
were, and deemed quite as indispensable, or we can hardly
believe that the nations and the individuals concerned would
keep the business alive as they have done, and are apparently
prepared to do. The great test of the utility of a product and
of its conser4uent chance of longevity, is the extent and the
tendency of its output. Judged by this standard, finished iron
has, of course, lost ground both relatively and absolutely, but
the loss is probably materially less than both producers and con-
sumers were prepared to expect. However this may be, we have
270 The Iron and Steel Magazine
collated and compiled the figures in the following table, showing
that in the United States and in Europe the output of finished
iron in 1904 was well over five million tons:
Rolled Iron. — Production in Leading Countries in Each of the
Last Two Years
1903 1904
United Kingdom (gross tons) 950.3^0 936,228
Germany (metrical tons) 819,832 765,197
France (metrical tons) 589,910 554,632
Belgium (metrical tons) 401,550 360,520
Russia and other European countries (met-
rical tons) 850,000 £00,000
Total 3,611,682 3,416,577
Add United States (gross tons) — 1,760,084
Grand total . — 5,176,661
The production of the United States in 1903 cannot be
stated, as the figures have not been collected, but it is not
likely to have differed greatly from that of 1904. The produc-
tion of finished iron outside of Europe, apart from the United
States, is not likely to have been very large, but it need hardly
be added that every civilized country does produce a larger or
smaller quantity of this material, and it is not at all improbable
that the total output of all countries will not be far short of six
million tons. This compares with the output of puddled iron
in previous years as under:
1880 1890
Tons Tons
United Kingdom 2,681,150 1,923,221
United States 2,332,668 2,518,194
Germany 1,106,800 1,454,131
France 1,056,160 823,360
Belgium 493, 000 506,957
Austria-Hungary 360,000 470,536
Russia 292,304 365,500
Sweden 231,143 278,70a
Total 8,553,225 8,340,599
The conclusion brought home by these figures is that the
finished iron trade of the world was quantitatively stronger
about 1880 than it has been at any later date, and that the
world's output at that period was al)Out three and one-half mil-
Mctalliiri^ical Xotcs and Comments 271
lion tons more than it is at the present time, the difference being
due. to the extent of about 1,700,000 tons, to the greater output
of the United Kingdom, and to the extent of 573,000 tons to the
larger production in 1880 of the United States, while Germany
has reduced her output in the interval by 342,000 tons.
The figures submitted are necessarily more or less imper-
fect, and in some respects are not quite com.parable, but even
so they come as near as it is possible for us to get. The uni-
formity of statistics for which we pleaded in our issue of last
week would enable us to get nearer to the actual facts. But
as the case stands at present, some countries make their returns
in puddled bars and others in finished steel, while a third group
make their returns partly in the one and partly in the other
description. A fourth group seek for " respite and nepenthe "
from the distractions of records and comparisons by making
no returns at all.
It is likely to be regarded by some of our readers as not a
little singular that a country usually so ready to publish records
of production as the United States should not for a number of
years past, until last year, have attempted to publish returns
of the output of wrought iron, Vet such is the case. The latest
records of the output of finished material under this head, pre-
vious to 1904, were collected for the year 1890, when the quan-
tity produced was 2,518,194 tons. This appears to have been
almost, if not quite, a maximum production. Previous to 1886
the maximum annual output was 2,360,649 tons in 1881; ten
years further back the output was only 1,292,396 tons. The
Statistical Report of the American Iron and Steel Association
for 1904 goes into details of the different descriptions produced to
only a ven,'' limited extent. All but about 5 per cent of the
total quantity produced is grouped under the general head of
merchant bars, skelp, spike rods, splice bars and other finished
rolled products. The more fully specified products are rails,
structural shapes, plates and sheets, nail plate and wire rods,
and their aggregate only amounts to about 98,000 tons, of which
■over 67,000 tons take the form of plates and sheets.
The point in this inquiry that is likely to possess the most
practical value for the iron trade as a whole is that of whether
a continuance of the recent decay of the wrought-iron industry
is to be expected, or whether that process has reached its ultimate
2y2 The Iron and Steel Magazine
limits. The answer is found to a certain extent in the fact that
although the process of decay has not of late years made the
progress that it did at an earlier period, it has not entirely ceased.
Nevertheless, the movement is still downward, and there is no
sign of its probable arrest.
Our own country has for several years past fluctuated
between 900,000 and 950,000 tons annually of puddled bar.
Germany has in the same period varied between 765,000 and
1,000,000 tons of rolled iron. France has similarly varied
between 554,000 and 650,000 tons. The future decay may be a
slower process than the past has been, but it is more than likely
to continue all the same. As the case stands to-day, the world's
output of steel ingots is about 35,000,000 tons annually, or
nearly six times the output of finished iron. The greatest output
of the latter product in any one year was less than a third of the
present output of steel. — " The Iron and Coal Trades Review,"
August II, 1905.
Conditions during the first half of this year
Our Actual Pig- ^^^^ unusually favorable for developing how
^°C ^ dtv much pig iron the United States can produce
under normal conditions. There was an ex-
cellent demand for iron, yet prices were not such as to bring
into play the scattered furnaces with high production costs,
so that conditions in the first half of this year were a nearer
approximation to those required for a full normal production,
and no more, than we have had for a long time. The statistics
of the American Iron and Steel Association show that pro-
duction of all grades and by all fuels was 11,163,175 gross tons,
exceeding by 1,455,808 tons the best previous half-yearly
record, made in the first half of 1903, with 9,707,367 tons. The
latest summary of total capacity is that presented in connection
with the last association directory, which gives the theoretical
capacity on June i, 1904, at 28,114,000 gross tons. Some of
the furnaces included in the total are furnaces which are un-
likely ever to produce, but have not been definitely abandoned.
The estimates of capacity in no case make allowance for relin-
ing, repairs and accidents. Accordingly, the actual capacity is
always below the theoretical capacity, and the record in the
first half of this year gives us a better idea than can usually
Miiailuri^iciil Xoics ami CoDuiicnts 273
be had of how much this cHfferonce is. The production was at
the rate of 22,326,350 tons per year, so that there is a difference
between the theoretical maximum and the practical maximum,
under normal conditions, of not more, and probably less, than
6,000,000 tons. It should be remembered that the dull period
of 1904 afforded unusual opportunities for anticipating neces-
sities of relining and other repairs, and it is a fact that the blast
furnaces of the United States entered the year 1905 under better
physical conditions than can usually be expected. On the other
hand, there was a net increase in capacity during the half year
of 748,500 tons, as shown by a recent association sunmiary.
This increased capacity contributed to some extent to the actual
production in the half year. Conditions are quite favorable
at the present time, and the outlook is clearly that the produc-
tion in the second half of this year will somewhat exceed that
in the first half, indicating a total production for the calendar
year 1905 of about 23,000,000 tons. It can hardly fall a half
million tons short of this, and is not likely to exceed it by a
million tons. One must go back only seven years, to 1898, to
find a production only about half as great, since in 1898 pro-
duction was 11,773,934 tons, yet that production broke all
previous records by more than 2,000,000 tons.
REVIEW OF THE IRON AND STEEL MARKET
August has hardly contributed as much improvement in the
iron and steel trade as did July, but the contribution has been
satisfactory nevertheless. The pig iron markets have been firm
and advancing, crude steel has been very scarce, and in most
finished lines there has been some improvement in demand.
While official prices of the heavier finished steel lines have
not changed, there has been some heavy contracting for certain
lines, and as the expiring contracts had been taken before official
prices had been advanced to the recently ruling level, the result
is a virtual advance, so far as the actual tonnage is concerned.
Since the middle of July there has been heavy contracting for
merchant steel bars, some of the contracts extending through
next June. In plates a large tonnage has been contracted for
1906 delivery, for shipbuilding interests on vessels already
contracted for.
There has been some improvements in all the lines which
have recently been dull, including merchant pipe, wire products,
sheets and tin plate.
Pig Iron. — Production of pig iron is at a slightly greater
rate than on July i and the turn seems to have come, with
prospects that production in the second half will closely ap-
proximate the unprecedented production in the first half,
officially ascertained at 11,163,175 gross tons. Sales have not
been as heavy in August as in July, but the market has gradually
assumed a firmer tone nevertheless. The United States Steel
Corporation is concluding the purchase of a large block of
Bessemer pig for early delivery at $i4-50' valley, equal to $15.35,
Pittsburg. We now quote the market as follows: F. o. b.
valley furnace: Bessemer, $14.50 to $i4-75; basic, $14.50 to
$14.75; No. 2 foundry, $14-25 to $14-50; g^^y forge, $13.75 to
$13.90. Delivered Pittsburg: Bessemer, $15.35 to $15.60;
basic, $15.35 to $15.60; No. 2 foundry, $15.10 to $15.35; gray
forge, $14.60 to $14.75. F. o. b. Birmingham: No. 2 foundry,
$11.50 to $12.00; gray forge, $10.50 to $11.00. Delivered
274
Review of the Iron and Steel Market 275
Philadelphia: No. 2 X foundry, $16.25 to $16.50; standard
grav forge. $14.75 to $15.00. Delivered Chicago: northern
No. 2 foundry, $16.25 to $16.50; malleable Bessemer, $16.50 to
S16.75. Freight: Birmingham to Pittsburg, $4.35; to Cin-
cinnati. $2.75; to Chicago, $3.65; to Philadelphia by water,
$3.50; by all-rail, $4.00.
Steel. — Late in July the market advanced a dollar a ton
on billets and 50 cents on sheet bars, to $24.00 for Bessemer
billets, $25.00 for long sheet bars and $25.50 for cut sheet bars,
all f. o. b. Pittsburg. Sales have not been large at these figures,
but have been sufficient to thoroughly fix the market, there
being no tonnage to be picked up at any lower prices. Forging
billets have had good sale at high prices, ranging from $26.00 to
$28.00, Pittsburg, according to tonnage and carbons. Ordinary
soft open-hearth billets are about $25.00. Wire rods are quoted
at $32.50. Pittsburg, and chain rods at $33.50.
Shapes. — New business in shapes cannot be placed with
anv of the large mills except for delivery next year. Some of
the smaller mills can make deliveries on new business within
two or three months. Some high prices are being paid for
small lots from dealers' yards, from 2.10 to 2.50 cents for carload
and larger lots. Regular mill prices remain at 1.60 cents for
beams and channels, 3 to 15 inches inclusive, angles 2 x 3 to
6x6 inclusive, and zees, 3-inch and larger; 1.65 cents for tees
3-inch and larger, and 1.70 cents for beams and channels over
15 inches.
Plates. — The shipbuilding interests have nearly all their
berths engaged for next year and have been contracting for
plates against this work. Most of the tonnage has already been
contracted for, plates for a few vessels Vjut lately contracted for
being now under negotiation. This business is all done at the
official basis of 1.60 cents, whereas nearly all the ship plates
for this year's consumption are on old contracts at 1.40 cents, so
that the mills are realizing results from the advances made six
months and more ago. No large orders for steel cars have been
taken recently, Vjut negotiations have been started on some large
contracts. Deliveries can only be promised for next year,
capacity being engaged fully for this year. The plate mills have
their product practically assured for the balance of this year by
actual specifications already in hand, and cannot make deliveries
276 The Iron and Steel Magazine
on new business for many months, except in the case of a few
small plate mills in the east. Official prices remain at 1.50 cents
for plates 14 inches wide and under, and 1.60 cents for plates
over 14 inches and not over 100 inches wide, tank quality,
quarter inch and heavier.
Merchant Bars. — About the middle of July a heavy con-
tracting movement set on for merchant steel bars, and to this
writing about 500,000 tons have been sold on contracts running,
in the case of agricultural implement makers through next June ; in
the case of most other manufacturers through December; and
in the case of jobbers thro\igh the third quarter. This business
was all done at the regular official price of 1.50 cents, base,
half extras, f. o. b. Pittsburg. Most of the contracts expiring
were at 1.30 cents. A few contracts are still running at 1.40
cents. In iron bars there has been a decided improvement, the
market advancing from $1.00 to $2.00 a ton. We now quote
common iron bars at 1.55 cents, Youngstown; 1.60 cents, Pitts-
burg; and 1.55 cents, Chicago. A few sales were made about the
beginning of the month as low as 1.45 cents, Chicago, but the
sharp advance in scrap prices stopped this business.
Sheets. — While there has been no sudden change in the
sheet market there has been a steady gradual improvement in
tonnage, which is now fairly satisfactory, while prices have not
responded materially. The leading interest is now operating
fully three fourths of its sheet capacity, and the independents
are doing about as well. On ordinary car loads for mill ship-
ment the market can be quoted at 2.35 cents for black and 3.40
cents for galvanized. No. 28 gauge, f. o. b. Pittsburg. On
larger lots it would be possible to do about a dollar a ton less,
and on very large lots, with favorable specifications and for
early shipment it is possible that $2.00 less than above prices
could be done. Corrugated roofing is $1.65 per square for
painted and $2.85 per square for galvanized. No. 28 gauge, in
carload lots. The tin-plate business has shown some improve-
ment, but not much, it being too early for consumers to commit
themselves freely against the fall canning crops. On very large
lots for early delivery the leading interest is giving a special
rebate of 15 cents a box, in addition to the regular- 5-cent rebate
from the official price of $3.55 for 100-pound cokes, f. o. b.
Pittsburg. This special rebate can be withdrawn at any time.
Rci'icw of tJic Iron and Steel Market 277
The independents have been making about the same net prices
on desirable orders. About half the leading interest's tin plate
capacity is in operation, and by the end of September it is
expected that nearly all its capacity will be employed. All but
one of the important independent tin-plate mills have been
operating full or part capacity during August.
Scrap. — Scrap prices have taken a sharp jump, more on
account of the limited visi1)le supply than because demand
from consumers has improved greatly. Early in August some
contracts for heavy melting stock were made at $14.00 and
$14.50 delivered to mills in the Pittsburg and valley districts,
but, on the short portion of some of these, dealers had to cover
at a loss, paying as high as $15.35 for scrap to apply on $14.50
contracts. The market is now $15.00 to $15.50 on small lots,
while it is doubtful if a large lot could be had at under $16.00,
the supply being limited. Cast borings have advanced sharply
and are quoted $9.50 to $10.00, delivered Pittsburg.
STATISTICS
Iron-Ore Production in 1904. — The annual statistics of
iron-ore production, as compiled by John Birkinbine, of Phila-
delphia, have just been published by the United States Geolog-
ical Survey. According to the report, the iron ore produced
in 1904 was 7,374,978 tons, or 21 per cent less than in 1903, the
figures for the two years being 27,644,330 as compared with
35,019,308 long tons. The continued decrease in tonnage since
1902, when a maximum production of 35,554,135 tons was
reached, is in contrast to the tendency before that time when
tonnage increased each year from 15,957,614 in 1895 to the high-
water mark in 1902. The report suggests that, 1902 and 1903
being phenomenal years, the decrease simply indicates a return
to normal conditions. The apparent consumption also fell off
about 12 per cent, aggregating 30,224,910 tons for the year.
The various ores making up the total output, red hematite,
brown hematite, magnetite and carbonate, were produced in
much the same proportion as last year. Magnetite alone was
produced in greater quantity this year than last and the car-
bonate ores fell off about one half. The production of red
hematite, which constitutes 86 per cent of all the ore produced,
fell off in the same proportion as the total production, the
amount being 23,839,477 long tons against 30,328,654 tons in
1903. Of the quantity mined Minnesota contributed over one
half, Michigan approximately one third, with AlaVjama, Wis-
consin and Tennessee following in order. The decrease in the
Minnesota production was most noticeable, the other states
maintaining about the same output.
The total quantity of brown hematite mined in 1903
(3,080,399 long tons), decreased in 1904 to 2,146,795 long tons,
a loss of 933,604 tons, or 30 per cent. Of the separate states,
Alabama had the largest output of this ore, followed by Vir-
ginia, West Virginia and Tennessee. The three states pro-
ducing the greater part of the magnetite were New York, New
278
Statistics 279
Jersey and Pennsylvania. In New York especially the output
was large o\\ injif to the activity in the Lake Champlain district,
and, as a result of this large production, the total for the year
rose to 1,638,846 long tons, an amount which exceeded the
tonnage of 1903 (1,575,422 long tons) by 63,424 tons.
The production of carbonate ores is restricted to Ohio and
Maryland and amounted to 19,212 tons in 1904 as compared
with 34,833 tons in 1903, a falling off of 15,621 tons, or 45 per
cent.
During the year 1904 there were produced in the United
States 370,118 long tons of concentrated ore, most of which
was magnetically separated. There were also made 68,189
tons of residuum, a zinc ore by-product used in the manufacture
of spiegeleisen.
The states producing ore last year were identical with those
of 1903 with the addition of Montana, and their respective out-
puts and corresponding values are given in the following table :
State Production Total Value Average Value
at Mines per Ton
Minnesota 12,728,835 $18,141,902 $i-43
Michigan 7,089,887 13,936,261 1.97
Alabama 3,699,881 3,744,35© i-oi
New York 842,303 1,885,291 2.24
Virginia and West Virginia . 550,253 951,478 1.73
Tennessee 499,949 1,126,687 2.25
New Jersey 500,982 566,813 1.13
Wisconsin 483,475 856,710 1.77
Pennsylvania 397,107 611,211 1.54
Georgia 293,802 356,950 1.22
Montana, Nevada, Texas,
New Mexico, Utah and
Wyoming 210,945 264,297 1.25
Colorado 150,972 429,856 2.85
North Carolina 64,347 80,440 1.25
Missouri 49,285 92,820 1.88
Kentucky 3 5, 000 3 5, 000 i.oo
Connecticut and Massachu-
setts 21,990 66,839 304
Ohio 15,672 21,829 1.39
Mar>-land 9,645 18,007 '1-87
Total 27,644,330 43,186,741 1.56
The Lake Superior region continued to maintain its position
as the leading ore producing district in the world, and the fol-
2 8o
The Iron and Steel Magazine
lowing table shows in brief the output of the five ranges which
comprise that region. The output of the Michipicotin Range,
which amounted to but 95,887 tons, is not included.
Range
1903
1904
Decrease
1903 to 1904
Rel Part
of District
Output
Marquette
Menominee
Gogebic .
Vermilion
Mesabi
Long tons
3,686,214
4,093,320
3.422,341
1,918,584
13,452,812
Long tons
2,465,448
2,871,130
2,132,898
1,056,430
11,672,405
Long tons
Per ct.
1,220,766
33
1,222,190
30
1,289,443
37-6
862,154
45
1,780,407
13.2
Per ct.
12.2
14.2
10.6
5-2
57.8
26,573.271
20,198,311 6,374,960
24
100
An inspection of the above table will show that all of the
ranges have fallen off from the production of 1903, with the
output of the Mesabi range, which has always been the largest
producer since its inception, decreasing the least.
A very satisfactory condition for the future has been made
known by the exploitation of iron-ore fields, the findings show-
ing reserves of ore in many states that exceed the total previous
output.
As a result of the slackness in the demand for and manu-
facture of pig iron, the quantity of iron ore imported decreased
from 980,440 long tons, valued at $2,261,008, in 1903, to 487,613
long tons, valued at $1,101,384, in 1904. This importation is,
in the main, from Cuba, Canada and Spain. The exports of
ore, on the contrary, have steadily increased. The demand
of the Canadian furnaces has steadily grown, and in recent years
also the exports have been enhanced by the German demand
for ores rich in iron and high in phosphorus for use in the manu-
facture of basic metal. Tlie 213,865 tons of ore exported in
1904, valued at $458,823, was an increase of 133,254 tons over
the 1903 movement.
Altogether the situation, while not booming at all, is very
substantial. " Iron Trade Review," July 20, 1905.
Iron Production in Belgium. — The output of the Belgian
blast furnaces for the five months ending May 31 is reported
as follows, in metric tons:
Statistics 28 1
iyo4 iQos Changes
Foundry Iron 43,77° 42,308 D. 1,462
Forge Iron 100,687 85,823 D. 14,864
Steel Pig 3q8,i5o 424,577 1- 26,427
Total 542;6o7 552,708 I. 10,101
The increase was wholly in steel pig, which includes both
Bessemer and l:)asic iron. Foundry and forge irons show de-
creases. The total increase was hardly expected, in view of
the trouble caused to the furnaces by the coal miners' strike
early in the year. " Engineering and Mining Journal," July 6,
1905.
World's Pig-iron Production. — Returns, final in some cases,
and close approximations in others, of pig-iron production in
1904 by countries which in 1903 produced about 95 percent of
the total, indicate a production in 1904 by such countries of
42,329,637 gross tons of 2,240 pounds. Assuming that the
countries for which 1904 returns are not yet available to have
made the same quantity in 1904 as in 1903, there is indicated
a world's pig-iron production in 1904 of 44,804,150 gross tons,
a decline of 1,090,563 tons from 1903, which held the record at
45,894,713 tons. In 1885 the world's pig-iron production was
iQ, 1 00,000 tons, or less than half as much.
RECENT PUBLICATIONS
Fabrication de UAcier, by H. Noble. 604 6JX lo-in. pages;
numerous illustrations. Vve. Ch. Dunod. Paris. 1905. Price,
paper covers, $8.75; bound, $9.25. One of the most conspicuous
features of this recent addition to the metallurgy of steel is the
complete absence of any reference or even allusion to the work
of other writers, from whom the author, nevertheless, necessarily
gathered much of the information contained in his book, nor do
we find any description of metallography and recent investiga-
tions into the treatment of steel. We also note the absence of
any reference to the cementation and crucible processes for steel
making, the work being limited to a detailed description of the
Bessemer and open-hearth steel processes. The author's treat-
ment of his subject is essentially practical, but little space being
devoted to considerations of a purely theoretical character. The
operations are described at length, and although they represent
French practice, there is much in the author's description which
should prove of interest to steel metallurgists of other countries.
The practical character and exhaustiveness of the book should
appeal strongly to metallurgists actually engaged in the produc-
tion of steel. It is the work of an experienced steel maker who
has an intimate knowledge of the operations he describes.
Modern Iron Foundry Practice, Part II, by George R. Bale.
194 5X7-in. pages; 95 illustrations. The Technical Publishing
Company, Manchester (England). 1905. Price, $2.50. — This
second part of Mr. Bale's well-known book deals chiefly with
machine molding and molding machines, physical tests of cast
iron, cleaning castings, shrinkage and distortion of castings and
foundry accounting. While the author's description necessarily
represents English rather than American practice, this book will
be read with interest and profit by American foundrymen and, in
general, by engineers interested in founding and in the physical
testing of cast iron.
282
Recent Publications 283
Concrete Steel, by W. Noble Twelvetrees. 218 5X7-111.
pages; illustrated. Whittaker & Co. London and New York.
1905. Price, 6s. — This is a short treatise on the theory and
practice of reinforced concrete construction. "The size of the
book and the clear and systematic treatment of the subject should
appeal to the busv engineer and architect who lack the time or
inclination to refer to more bulky volumes or to scattered articles.
Chemical and Metallurgical Handbook, by ]. H. Cremer and
G. A. Bicknell. 406 4X6-in. pages, full seal, gilt edges. Cleve-
land, Ohio. 1903. Price, $3.50. — This excellent handbook is
of such value to metallurgists and chemists that we feel justified
in noticing it in our columns two years after the publication of
the last edition. The book includes useful chemical tables, min-
eralogical tables, tables of heat, metallurgical tables, weights and
measures and mathematical tables. The information which they
convey so readily is of such assistance to working chemists and
metallurgists that we feel confident that very few would fail to
procure a copy of the book after having examined it. Our only
averse criticism refers to the table of melting points of metals on
page 181, in which some of the temperatures given are not in
accordance with the results of recent and accurate experiments.
The melting point of aluminum, for instance, is given as 850° C,
while it should be in the vicinity of 650°; the melting point of
gold is given as 1147° C., while it is very close to 1064°. The
melting point of nickel which is given as 1450° C. is also con-
siderably too low.
Outlines 0} Industrial Chemistry, by Frank Hall Thorp, in-
cluding .4 Chapter on Metallurgy, by Charles D. Demond. Second
edition, revised and enlarged. 618 52X8^-in. pages; 116 illus-
trations. The Macmillan Company. New York. 1905. Price,
$3.50. — The first edition of Professor Thorp's book was pub-
lished in 1898 and received with much favor, especially as a text-
book for the teaching of industrial chemistry. In this new edi-
tion the text has been considerably revised and enlarged, this
being made necessary by the important advances in the chemical
industries since the appearance of the first edition. The addi-
tion of a chapter on metallurgy adds much, we think, to the
value of the book, especially for those schools where metallurgy
284 The Iron and Steel Magazine
is merely taught as a part of the course in industrial chemistry.
The author divides his subject into three parts: (i) inorganic
industries, (2) organic industries and (3) metallurgy. The
more important "industrial chemical processes are briefly, sys-
tematically and clearly described, and numerous references are
given for those who desire to make a more exhaustive study of
any process. This second edition of Professor Thorp's excellent
work will not fail to be warmly welcomed. by those interested in
the teaching of this important and growing subject, and the pub-
lishers are to be congratulated for the evident care with which
the book has been prepared.
James Watt, by Andrew Carnegie. 241 6X9-in. pages.
Doubleday, Page & Co. New York. 1905. Price, $1.40. — The
contents of this very interesting, instructive and suggestive book
might be divided into two parts : first, the narrative proper of
Watt's life, simply and attractively told, and, secondly, the
author's comments on literary, philosophical, economical and
ethical questions as occasions arise in the narrative, and which
possibly form the most fascinating portion of the book. These
comments are the thoughts of a thinker and of a humanist en-
dowed with a power of expression possessed by few, and of
extremely rare occurrence in men having devoted their lives to
industrial pursuits and money making.
Production and Use of Petroleum in California, by Lewis E.
Aubur\^ state mineralogist. 230 6X9-in. pages; 64 illustra-
tions; paper covers. California State Mining Bureau. Sacra-
mento. 1904. Price, $0.75. - — In his letter of transmittal Mr.
Aubury states that this bulletin (No. 32) has been prepared by
Mr. Paul W. Pratzman, engineering chemist, chiefly from his own
observations. The work of the bulletin has extended over two
years and the greatest care has been exercised in the collection of
accurate data concerning the subject treated. The aim of the
bulletin is to describe the conditions under which petroleum is
produced, the amount, source and character of production of
same, and to outline some of the more important ways in which
the output is consumed. From the reading of this bulletin the
general public should obtain a clear knowledge of the conditions
existing in the oil industry at the present time.
Rccctit Publications 28
:>
Geology oj Wcsicrn Ore Deposits, by Arthur Lakes, late pro-
fessor of geology in the Colorado State School of Mines. New
edition, rewritten and enlarged. 415 6X8-in. pages; 300 illustra-
tions. The Kendrick Book and Stationery Company. Denver.
1005. Price. $2.50. — The author states that his object has been
to illustrate the principles of mining geology rather than to dilate
on any particular rich or noted mine. Colorado mines chiefly are
described, and most of the illustrations are sketches from the
author's pencil taken in the field. A glossary of some scientific
and other terms used with mining field is appended.
Xotes and Questions in Physics, by John S. Shearer, assist-
ant professor of physics, Cornell University. 281 5^X82-in.
pages; 202 illustrations. The Macmillan Company. New York.
1904. — This book contains 1,497 problems carefully selected to
illustrate physical phenomena and calculations, including the
mechanics of solids, liquids and gases, heat, magnetism, electri-
city, sound and light. Some useful physical and mathematical
tables are also appended. The author claims, undoubtedly on
good ground, that students continually complain of their inability
to solve simple problems in physics, which is a clear indication
that the fundamental principles are not fully grasped, and that
concrete, practical examples are needed to bring out the sense of
the terms used. In order to suggest methods, a few typical cases
have been fully solved in the text. It is somewhat regrettable
that the answers to the problems have not been included, because
this omission decreases the value of the book for the private
student. The book is well printed and illustrated and attrac-
tively bound.
Western Mill and Smelter Methods oj Analysis, by Philip H.
Argall. 124 5^-X7^-in. pages; illustrated. Industrial Printing
and Publishing Company. Denver, Colo. 1905. Price, $1.50. —
This book consists of a description of the methods of analysis used
in lead and copper smelters and in cyanide mills in the Western
states. The chapter on the cyanide process, including a detailed
description of the daily work of the chemist, should prove of
value to those interested in that process. It is claimed by the
publishers to be the most practical synopsis of the subject ever
written.
2 86 The Iron and Steel Magazine
Moody s Manual of Railroads and Corporation Securities,
igoj. 2,642 6^X9-111. pages. Moody Publishing Company.
New York. 1905. Price, cloth, $10.; full leather, $12.^ This
new edition of " Moody's Manual of Railroads and Corporation
Securities" is so comprehensive and complete that the claim made
by the publishers that the book is superior in every department
to that of any publication in the country is a statement that
cannot be repudiated. The book is four inches thick and weighs
over twelve pounds. It contains over three million words and
covers the entire field of corporation investments. There are
ten sections to the volume, and each section has been prepared by
its own special experts, who have made it their entire work to
make the book complete and up-to-date in every possible sense
of the word.
It is probably the most accurate and complete steam rail-
road reference book in existence; it is the only electric traction
book that undertakes to cover the entire field in this compre-
hensive way; it is the only reference book containing a section
on gas and electric light companies of the country ; it is the
only book containing a section on water-supply companies; it
is the only reference book containing a section on telephone and
telegraph companies; it is the only book containing a section
on industrial and miscellaneous corporations which is in any
sense uniform and complete ; and it is the only investment man-
ual containing reports on mines and oil corporations. Further-
more, it is the only book in the United States which contains
anything like complete statements of banking and financial insti-
tutions, showing the essential facts of interest to investors in
bank or trust company stocks. It is also the only book contain-
ing complete and up-to-date lists of the members of the twenty-
five stock exchanges of the cities of the United States.
The publishers state that since the publishing of the first
edition, which appeared in 1900, the growth in pages has been
over 1 40 per cent ; the growth in the size of this volume over the
first edition is 100 per cent; the increase in quantity of contents
as compared with the first edition is over 800 per cent ; and the
increase of the current edition in circulation over that of 1900 is
over 400 per cent.
PATENTS
RELATING TO THE METALLURGY OF IRON AND STEEL
UNITED STATES
793,110. Gas-Purifer. — Edward A. Uehling, Passaic, N. J.
793.137- Electric-Magnetic Ore Separator. — Erich Langguth,
Euskirchen, Germany.
793,268. Metal-Annealing Furnace. — Darwin Bates and
George W. Peard, Huyton, England.
793,305- Gas-Producer. — Ernst Korting, F. Paul, Jr., Peeks-
kill, N. Y., assignor, by direct and mesne assignments, to Henry Amling,
Jr., New York, X. Y
793,350- Tuyere for Forges. — John Christian, Hydraulic, Colo.,
and Louis S. Judd, Oak Park, 111.
793,377- Casting Plant. — Joseph G. Johnston, Detroit, Mich.,
assignor to American Car and Foundry Company, St. Louis, Mo.
793,380. Blowing-Engine. — Albert T. Keller, Wilkinsburg, Pa.
793,425. Apparatus for Forming Pipe-Molds. — Jacob K. Dim-
mick, Philadelphia, Pa.
793,467. Tuyere-Iron. — Soren P. Petersen and Wilson Toups,
Patterson, La.
793,501. Annealing Furnace. — Carl Bechstein, Cannstatt, Ger-
many.
793,544- Apparatus for Purifying Blast-Furnace Gases. —
Walter Schwarz, Dortmund, Germany.
793,554- Gas-Producer. — William Viggers and Rowland Z. Ball,
Durango, Mexico.
793,745- Means for Removing Dust from Gases. — John Shields,
Willesden Green, England.
793,806. Furnace-Charging Apparatus. — George Schuhmann,
Readinrr^ Pa., assignor to Reading Iron Company, Incorporated,
Reading, Pa.
793,852. Furnace-Charging Box. — Cameron C. Smith, Pitts-
burg, Pa.
793,877. Ingot-Delivery Car. — Thomas James, Braddock, Pa.
794,000. Process of Purifying Gas. — Adalbert W. Fischer,
Philadelphia, Pa., assignor to Schutte & Koerting Company, Philadelphia,
Pa.
794,153. Utilization of Flue-Dust. — Charles S. Price, West-
mont, Pa.
287
28S
The Iron and Steel Magazine
794,201. Centrifugal Gas-Purifying Apparatus. — Edward
Theisen, Baden-Baden, Germany.
794,386. Furnace-Charging Apparatus. — Ralph Baggaley,
Pittsburg, Pa.
794,391. Furnace-Charging Device. — John J. Boax, McKees-
port. Pa., assignor to National Tube Company, Pittsburg, Pa., a corpora-
tion of New Jersey.
GREAT BRITAIN
1 1 ,43 7 of 1904. Briquetting Iron Ores. — E. Goldschmid, Frank-
fort, Germany. Heating friable iron ores in a water-gas furnace in order
to obtain coherent masses suitable for smelting.
12,367. Gas-Producer. — Alfred B. Dufif, Pittsburg, Pa.
13,594 of 1904. Dust Collecting. — B. H. Thwaite, T. J. Denny
and R. E. Commans, London. Collecting the dust given off at the work-
ing faces of mines by means of powerful suction pipes.
411 of 1905. Smelting Iron Ore. — J. Gayley, New York. In
smelting iron ores with a dried-air blast, using much less coke than is
theoretically required, the explanation of this economy not being quite
clear.
SIR LOWTHIAN BELL
SEE PAGE 354
Ag 9
The Iron and Steel Magazine
" Je veux au mond puhlier
d'une plume de fer sur un papier d'acter."
Vol. X October, 1905 No. 4
ELECTRIC STEEL*
By F. W. HARBORD
'THHE great interest which the manufacture of steel in the
electric furnace has aroused, both amongst manufacturers
and engineers in this country, and the fact that there is already
one electric furnace in Sheffield and that it is reported that
another Sheffield company has acquired the exclusive patent
rights of the well-known Heroult process for the British Isles,
make it important that we should consider the possibilities of
electric smelting relative to steel manufacture in England.
While, on the one hand, the extravagant claims urged on
behalf of electric smelting — that it will revolutionize the manu-
facture of structural steels as at present made by the Bessemer
and open-hearth process — may be dismissed as nonsense, the
attempts, on the other hand, to prove that it cannot compete
with the crucible process in the manufacture of tool steels, or
the open-hearth furnace for many of the higher class steels inter-
mediate between these and common structural steel, may equally
be disregarded. The truth lies between these two extremes,
and the manufacturer who realizes this and takes advantage
of the great possibilities which the electric furnace offers to
meet very many of the special steel requirements of to-day, and
who does so with judgment and knowledge, will, without doubt,
be in a most exceptional position, not only to meet foreign com.-
petition, but to more than hold his own against his British com-
petitors. Since the Canadian Commission, visited Europe last
♦"The Times Engineering Supplement," August 2, 1905.
290 The Iron and Steel Magazine
year, rather more than a year has elapsed.* During this time
very considerable quantities of electric steel have been made
both in Sweden and in France, and have been used with most
satisfactory results for all classes of tools and cutlery, and for
various other purposes for which the highest class crucible
steel was formerly employed, confirming in every way the con-
clusions of the Commission that " steel equal in all respects to
the best Sheffield crucible steel can be made." Considerable
quantities of this steel have been supplied to Sheffield firms, who
have thus been able to convince themselves of its exceptionally
high quality, and it now only remains for our Sheffield people to
make the steel for themselves, rather than import it. The
manufacture of crucible steel for tool purposes, important as it
is to the country, owing to the world-wide reputation for quality
which it has acquired, is, however, only one comparatively
small branch of our great steel industry; and perhaps the most
important question is, to what extent electric smelting can be
employed for the manufacture of the numerous classes of steels
between this and ordinary Bessemer, or open-hearth steel.
We import annually very large quantities of Swedish Bes-
semer steel for tube blanks, for the solid drawn tube trade, and
for other purposes too numerous to mention; again, large quan-
tities of Swedish pig irons are imported for use in our open-
hearth furnaces for the manufacture of special qualities of high-
class steel for large forgings, axles, tires, special wire and other
purposes, and in many cases steel of the required composition
can only be made by using, either entirely or in part, these very
high-priced pig irons. Another very important branch of the
steel trade is the production of dynamo steel of exceptional
purity and low" hysteresis, and in this direction the electric fur-
nace promises great things, as steel of the greatest purity, low
in carbon and manganese, can readily be produced. If we add
to these the manufacture of all kinds of ordnance, armor plate,
projectiles, rifle, bayonet and other high-class steel, we see that,
without attempting to compete with Bessemer or ordinary open-
hearth structural steel, there is an immense field open to the
electric furnace. Numerous experiments have shown that elec-
tric steel is not only extremely pure, but it is also exceptionally
* A review of this report of the Commission appeared in " The Times
Engineering Suj)i)lement," March 8.
Electric Steel 291
homogeneous, and this is a most important point in the manu-
facture of large steel castings. When it is remembered that, for
special purposes, castings, sometimes of fifty to sixty tons, have
to be made by mixing the contents of a number of crucibles not
containing more than i cwt. each, the advantages of being
able to make steel equal in all respects as to quality, in quan-
tities of fifteen tons and possibly more, will readily be apparent.
If steel to satisfy the exacting requirements of the highest
class of tool steel can be produced, there can be no question
as to the production of steel of a quality suitable for what we
mav term medi\im-class steels, and it then becomes simply a
question of cost, and whether the electric furnace can compete
in this respect with Swedish Bessemer steel, or steel made from
Swedish pig iron, or steel of specially selected English brands.
In the electric furnace of the resistance type, which may
be said to be represented by the Heroult and Keller furnaces,
the highest class steel can be made from ordinary English scrap,
such as rail ends, but against the saving effected in this direction
has to be set the cost of the electric energy required. The elec-
tric furnace, even under the best conditions, is not a cheap
melter, but as a refining furnace towards the end of the opera-
tion, when a very high temperature is required, it is far more
efficient; it therefore seems probable that the future develop-
ment of the electric furnace will be in combination with some
form of continuous open-hearth process, in which molten pig
iron is first converted into what we may term " molten scrap
steel " in a gas-fired furnace, and then transferred in the molten
state to the electric furnace for final purification. By this
means the additional cost over ordinary open-hearth steel would
be comparatively small, the melting and preliminary refining
having been done in the gas-fired furnace, and the electric
furnace being employed only to do the final refining at such high
temperatures as those at which it alone is able to work most
efficiently and economically.
The design of the Heroult furnace, so far as the general
construction is concerned, is particularly well adapted to work
in combination with an open-hearth tilting furnace, and if,
instead of charging cold scrap, or even molten pig iron, con-
verted metal were charged on some such lines as suggested, a
steel superior to best Swedish steel, or steel made from vSwedish
2Q2 The Iron and Steel Magazine
pig iron, should be obtained at a less cost. Given a large output,
so that labor costs are reduced to a minimum, the |jrice at
which such a steel could be produced would, no doubt, induce
many manufacturers to employ it for purposes for which at
present they are content to use inferior steel, and thus, it would
soon create a demand for high-class material, apart from that
already existing. It is not suggested that a simple refining of
ordinary steel in this way would be sufficient for the production
of the highest class of tool steels. For the production of these,
it would no doubt be necessary to carry on the operation in
the electric furnace in a way similar to that employed at La
Praz, at a considerably greater cost as to expenditure of electric
energy, time and labor; but in these cases the process is not
competing with the open-hearth method, but with the crucible
process, in which, although the output may be comparatively
small, there is a much greater margin as regards cost of produc-
tion, and the question of a pound or so a ton is of no great
consequence.
There are two other types of furnaces, known as (i) the
induction furnace and (2) the arc furnace, which are now com-
peting with the resistance furnace for the favor of the English
steelmaker. The former is represented by the Kjellin furnace,
which has been at work for several years in Sweden, and the
latter by the Stassano furnace, which has been at work for a
considerable time in Italy. The Kjellin furnace is quite dis-
tinct both in principle and construction from the Heroult fur-
nace, whilst the difference between the principle of the latter
and arc furnaces generally is not so clearly marked, and they
merge one into the other. But this classification will be suffi-
cient for practical purposes. In general arrangement, and also as
regards electrical and other details, the Stassano furnace is
totally distinct from the Heroult, and it was primarily designed
for the direct smelting of iron ore rather than for steel making,
although it has been producing steel most satisfactorily for some
time. From a practical engineering and metallurgical stand-
point, however, there can be no doubt that the Heroult furnace is
far better designed to meet the general requirements of the steel
manufacturer than either the Stassano or the Kjellin furnace.
It is understood that a furnace of the latter type is already
^t work in Sheffield, and there can be no question as to the
Electric Steel 293
quality of the steel produced, provided high-class material, such
as Walloon scrap, is used for its production. In Sweden, where
a furnace is attached to works producing this high-class scrap,
probably this furnace is as good, and may, under such conditions,
be even better than the Heroult; but the objection to it under
English conditions is its lack of adaptability, both as regards
the materials which can be used, and in any variation in design
to suit the conditions of our practice. In reality it is a large
melting crucible, and to get the highest class of steel it is neces-
sarv, just as in the crucible process, to charge the purest mate-
rials, as the amount of purification which takes place during
the operation is practical^ very small. On the other hand, the
Heroult process can deal with ordinary English scrap or pig
iron, and by the repeated addition of suitable fluxes to form
new slags, the impurities can be removed, so that a final product
is obtained quite equal, if not superior, to much that is made
from Swedish materials in a crucible.
That steel made in an electric furnace should possess supe-
rior properties to steel of similar composition produced either in
a Swedish Bessemer converter, or in an open-hearth steel fur-
nace, may seem at first to be claiming a great deal, but such
appears to be undoubtedly the fact, and this is due probably
to its production in what may be regarded as a practically
neutral atmosphere, under conditions in which the occlusion of
gases and over-oxidation is reduced to a minimum.
It is frequently urged that the cost of electric energy in
this country makes the production of steel in anything like
quantities a commercial impossibility; but with electric energy
at ;^io per kilowatt year, at which price it can be produced under
favorable conditions from coal, and by using the gas furnace
for the melting, and the electric furnace only for the final opera-
tion, the difference in cost, as regards electric energy, will prob-
ably be more than met by the lower price of our raw material,
and our proximity to markets for the sale of the finished product.
When the irregularity in supply, due to the change of seasons,
and the generally inaccessible position and remoteness from
sources of supply, and from markets for the sale of the finished
product are taken into consideration, the much-talked-of cheap
production of electric energy from water power will often be
found to be more apparent than real.
2 94 The Iron and Steel Magazine
THE GALBRAITH ELECTRIC IRON AND STEEL
FURNACE *
\ SUCCESSFUL treatment of iron sand would be an important
step in advance. Iron sands abound in many localities.
The term is used both for grainy magnetite — a combination of
iron oxides corresponding more or less to the formula FcgO^ —
and for the ore known to the mineralogist as menachanite (from
Menachan, in Cornwall) and by other names, an oxide of iron and
titanium, with 30 and more per cent of the latter. Both the
ores are either too compact for direct reduction, or so finely
^divided that they trickle through the furnace, and most at-
tempts to transform these ores into briquettes have failed.
Grondal and Mathesius are said, however, to have succeeded,
each in his own way, in making suitable iron-sand briquettes.
In any case, however, the new process which the Galbraith Iron
and Steel Company, Limited, of Auckland, New Zealand,
demonstrated at the Brush Electrical Engineering Company's
works, at Loughborough, Leicestershire, on Wednesday last,
would deserve close attention. The main features of this process
are the following : ' The iron sand mixed with carbon is passed
through a furnace so as to fall as a shower in a zigzag path
through a number of carbon grids heated by the current, until,
melted and reduced, it arrives in the receptacle below. The
whole furnace is sealed, and open only at the feed-hole above,
so that no air can enter, otherwise the grids or incandescents,
as the inventors, Messrs. J. K. Shirrefit' Galbraith and William
Steuart, call them, would be burned. The charge is further
met on its downward path by reducing gases. Low potential
currents are utilized. Neither flux nor fuel, in the ordinary
sense, is needed, and the iron or steel, which collects in the
receiver in the shape of beans, is said to be free of the titanium
contained in the ore. An admixture of i per cent would not be
objected to by metallurgists in most cases; often a higher per-
centage of titanium could be tolerated, and for certain steels
some titanium is considered a desirable constituent.
The raw material comes from Taranaki Bay, in New Zea-
land, where it covers miles of the beach. This iron sand has,
* " Engineering," July 21, 1905. Abridged.
The Galbraith Electric Iron and Steel Furnace 295
so far, practically no value, and if the inventors do not accomplish
more than to transform the ver}^ fine black powder into beans
suitable for Bessemerizing or other treatment, they will have
achieved an object on which vast sums have, so far, been spent
in vain. The fine sand is passed through magnetic separators.
It is verv pure, and the impure and the purified ore shown
differed hardly in appearance. We give analyses of the ore
and of the steel resulting from another less pure sample of ore
taken from a breakwater:
Ore Steel
Per cent Per cent
Peroxide of iron 67.04 Carbon by combustion . . • 2.891
Protoxide of iron 30.17 Silicon 0.201
Manganese peroxide 0.22 Sulphur 0.189
Aluminum oxide 0.16 Phosphorus o-453
Silica 0.50 Arsenic nil
Calcium and magnesium .... traces Manganese o-i37
Titanium 1.6 Copper 0.24
Undetermined 0.31 Iron 96.095
Titanium nil
100.00 100.206
The titanium percentage may rise to 4 per cent. The
titanium is supposed to pass into the slag, but there are hardly
any constituents to form a slag with. The ore was supposed to
be free from phosphorus; the material just now worked on does,
however, contain some phosphorus.
The ore is mixed with coal dust and heated, for a few min-
utes only, in a crucible. This roasting, or preheating, would,
in a commercial plant, be performed in a rotatory kiln or in some
other less primitive way. The furnace itself is built up of a
framing of graphite, holding horizontal grids, likewise of graph-
ite; the grid-bars form obtuse-angled roofs. Four of these
grids are confined in one tier, and three tiers are arranged above
one another, so that the charge shower falls, in descending,
over twelve bars in succession and finally into the receiver —
an iron box — below. Between the bars, or incandescents, in-
terceptors are to be placed; these interceptors are of the same
shape as the incandescents, but they are not so thick, and their
bars are closer together. In this demonstration furnace only
one intercepter was used, at the very top of the furnace ; through
this intercepter grid the charge trickled down in three streams.
296 The Iron and Steel Magazine
The incandescents and intercepters are made by the Morgan
Crucible Company of Battersea.
Each incandescent, it will be understood, forms a rec-
tangular horizontal frame, against the front and the back of
which iron bars are pressed. In the furnace which formed the
subject of Wednesday's experiment, a good contact between
the front and the iron was secured with the aid of weighted levers ;
this pressure forced the backs of the grids against the bars at
the back. The current leads, copper bars, or ribbons, were
screwed to the iron; there were twelve circuits, one for each
grid. Mr. Gardner, who assisted in the experiment, put in
about a pound of charge per minute, and a current of 100 kilo-
watts at 18 volts was stated to be used. The waste of power
and heat in this skeleton furnace was, of course, great. But
furnaces of this type are to be constructed with walls of bauxite,
and it should be possible to cool the electrodes with the aid of
water-jackets. No figures as to the economy of the process can
be given under the circumstances. Through peep-holes, closed
by mica, the metal could be seen to trickle down into the re-
ceiver in big drops, which united to ordinary size beans, and
further fused together, as the receiver was not cooled. Speci-
mens of steel obtained in the furnace were exhibited.
We should only like to make one remark as to the power
consumption by such melting or smelting operations. Ruthen-
burg has claimed to melt his ore at an expenditure of heat en-
ergy below the theoretical figure. That is not inconceivable,
for it would suffice to melt a portion of the ore which would
cake together with other particles. In actual practice the
charge would pass hot direct from the rotatory kiln into the
furnace, and the final product would be remelted in a crucible
or converter. The carbon percentage, about two in demon-
stration samples, could more or less be regulated, if not in the
furnace, certainly in the converter. The furnace used for the
demonstration was certainly very experimental ; but it did some-
thing so far not accomplished, and it may perform its work
economically. Though it is too early yet to express any opinion
on this point, we heartily wish the inventors success. Mr.
Albert Robins is secretary of the Galbraith Iron and Steel
Company, of Auckland, whose London offices are at the Cannon
Street Hotel, E. C.
The Melting Points of Slags 297
THE MELTING POINTS OF SLAGS AND OTHER MEMBERS
OF THE SERIES SiO.-Al.O^-CaO *
By CLIFFORD RICHARDSON, Director New York Testing Laboratory, Long Island
City, N. Y.
'T^HE presentation of a paper before a recent meeting of the
■^ Iron and Steel Institute, May, 1905, by O. Boudouard,
entitled, " Experiments on the Fusibility of Blast-Furnace
Slags, "t has again attracted attention to a subject the litera-
ture of which has been quite abundant in recent years. To
review this in its entirety would be burdensome, but it will
be worth while to call attention to the fact that many of the
melting points and the conclusions drawn therefrom, which are
contained in the paper of Boudouard, and in the " Silikat-
schmelzlosungen " of J. H. L. Von Vogt,J are very seriously
in error, as shown by the work of Day and Allen on " The Ther-
mal Properties of Feldspars," § and by the investigations of
the writer on " The Constitution of Portland Cement from a
Physico-Chemical Standpoint." ||
If we examine the data given in the second volume of Von
Vogt, published in 1904, we shall find a resume of the melting
points of a large number of native minerals as determined by
prominent investigators: Doelter with the pyrometer, Joly and
Cusack with the meldometer, and Brun with Seger cones. To
select a few examples, it appears that wollastonite was found
to have a melting point of about 1245°, augite in the neighbor-
hood of 1160°, the amphiboles from 945° to 1135°, the olivine
group between 1065° and 1400°, while anorthite was found by
Doelter to have a fusing point of from 1124° to 1190°, and by
B,run of 1500°. With the exception of the latter determination
the results are all suspiciously low. That this suspicion is not
* Received August 16, 1905.
t Abstract. "The Iron and Steel Magazine," Boston, Mass., Vol.
X, No. I, Jvtly, 1905, p. 53.
X " Die Silikatschmelzlosungen," J. H. L. Von Vogt, Christiania, 1904.
§ " The Isomorphism and Thermal Properties of the Feldspars,"
Arthur L. Day and E. T. Allen. " The American Journal of Science,"
Vol. XIX, February, 1905, and publication of Carnegie Institute, Wash-
ington, D. C, 1905.
il " The Constitution of Portland Cement from a Physico-Chemical
Standpoint," Clifford Richardson, Long Island City, N. Y., 1904.
298 The Iron and Steel Magazine
unfounded can be seen from the results of the very careful
examinations of Day and Allen on the melting point of
anorthite, which was found by them to be 1532°. If we refer
to the tri-axial diagram given in Boudouard's paper, we find
that the locus of anorthite would indicate a melting point of
1500°.
In the discussion of the total heat of fusion of the various
minerals, Vogt assumes its melting point to be 1200°. His de-
termination, and that of Doelter, are over 300° out of the way.
Brun and Boudouard are nearer the correct figure, and as close
as could be expected with the means which they employed
for the determination. He similarly assumes equally low
figures, in the neighborhood of 1200°, for the fusing point of
diopside, microcline, akermanite and other silicates. From
the writer's observation these figures are as much out of the way
as the anorthite determination. Boudouard is as much in
error in his statements as to the melting points of the more
basic compounds as is Vogt in regard to anorthite and the other
natural silicates.
On examining Boudouard's tri-axial diagram it will be
found that, while his figures for the melting points of the alumi-
nates of lime and silicates of alumina of .various degrees of
basicity are fairly correct, considering the method which he em-
ployed for determining them, those which he gives for the sili-
cates of lime are so far out of the way that it would seem that
he could not have had these materials actually in hand. It is
a well-known fact to those who have studied the basic silicates
of lime from the point of view of their relation to the constitu-
tion of Portland cement that di- and tri-calcic silicate can only
be fused at a temperature above the melting point of platinum.*
It has also been plainly shown by the writer that silica, alumina
and lime, when combined in the proportions necessary for the
production of a Portland cement, while fusing at a lower
temperature than the simple silicates, melt at a higher tem-
perature than platinum. According to Boudouard's tri-axial
diagram the compounds included in the area correspond to
the composition of pure Portland cement and melt between
* " The Constitution of Portland Cement," Clifford Richardson.
" Cement," Progress Publishing Company, New York, 1904, Vol. V,
No. 3, p. 124.
TJic Melting Points of Slags 299
1450° and 1470°, while di-calcic silicate is shown with a melting
point of 1460°, and the tri-calcic silicate with a melting point
of 1500°.
It appears, therefore, that Vogt is entirely out of the way
in his conclusions in regard to the melting point of the feldspars,
as shown by Day and Allen, while Boudouard's data in regard
to the more basic compounds are equally incorrect. While the
writer cannot say that the melting points for the slags which
Boudouard gives are equally erroneous, it is plain that a cloud
is cast upon all his statements from the fact that he is so far
in error in regard to the compounds, the fusing points of which
are well known.
To those who are interested in the accurate determination
of the fusing points of slags and silicates, the work of Day and
Allen is recommended as showing the great care and refinement
of manipulation which is necessary for such determinations, it
being the only work which bears upon its face evidence of the
highest degree of accuracy.
Boudouard's methods are open to very serious criticism.
He determines the melting point by a comparison of the behavior
of the material to be examined, when made into a proof cone of
the exact dimensions of the Seger cone, with Seger cones of
known melting point. He states that the Seger cones are made
from Fontainebleau sand, white marble and calcined alumina
made from ammonia-alum, ground to such a degree of fineness
as to pass through a No. 100 sieve (with 1,370 holes to the square
centimeter), but not through No. 150 (with 3,080 holes to the
square centimeter). It is a well-known fact, which has been
shown by various investigators of the constitution of Portland
cement, that materials as coarse as this do not readily attain
•equilibrium except on exposure for a very long period of time
to temperatures above that of the heat of formation. The
di- and tri-basic silicates of lime can only be formed in a state
of equilibrium when the silica and lime of which they are made
have been reduced to such an impalpable state that they will re-
main suspended in water for a very considerable length of time,
and the same may be said of the solid solution of aluminates in
•silicates which form Portland cement. In the case of coarse ma-
terials the resulting compounds are always in metastable equi-
librium, or are found, under the microscope, to have an entirely
300 The Iron and Steel Magazine
different structure from those produced from the impalpably
fine powder.*
In view of these facts, the Seger cones prepared by Bou-
douard can hardly have furnished a proper indication of the
points of fusion of the more basic compounds which he has.
represented in his tri-axial diagram. It is plain, too, that exact
determinations can only be made on the lines followed by Day
and Allen, and that all melting points of members of the series,
SiOg-AlaOg-CaO which have been published are far from the
truth, in most instances, and open to very serious criticism
DESCRIPTIVE METALLURGY OF IRON AND STEEL.t 11
By SAMUEL GROVES
{Continued from page IQQ)
Ores in Nature
TT is a remarkable fact that in nature things which are of vital
importance to the existence and evolution of man are uni-
versal — either in adaptability or diffusion. For example,
wheat, " the staff of life," is the only cereal which can be grown
in every clime. Iron, " the king of metals," is found in every
part of the globe.
Universal Diffusion. — As W. Mattieu Williams has said,
we cannot dig up a spadeful of earth without finding oxide of
iron. No geological formation is free from it. It is found in
the ocean, in mineral springs, in the red blood, in the atmosphere,
in the very heavens above. So rarely is it absent from the soil
that a bed of sand free from it is as valuable as a gold mine.
Glass-workers have no little difficulty in getting such supplies.
Clay free from iron oxide is of great value to the potter. The
* " The Constitution of Hydraulic Cements," S. B. Newberry and
W, B.Newberry. " The Jour. Soc. of Chem. Ind.," 1897, Vol. XVI, p. 887.
" The Constitution of Portland Cement," Clifford Richardson.
" Cement," Progress Publishing Company, New York, 1904, Vol. V,
No. 3, p. 124.
t " The Canadian Engineer," September, 1905.
This article is copyrighted in the United States and is reproduced
here through the special permission of Mr. vSamuel Groves, editor of " The
Canadian Engineer."
Descriptive Metallurgy of Iron and Steel 301
prevailing reddish brown color of the earth is due to iron. Snow-
is looked upon as an emblem of purity, and yet within the Arctic
Circle one cannot evaporate a handful of snow without leaving
behind a sediment containing darkish particles which jump
to a magnet.
It does not need high powers of inductive reasoning to
perceive in this phenomenon of the universal diffusion of iron
a wonderful example of design in nature. Without metals man
would have remained a savage. Louis Figuier has luminously
enforced this view in his " Primitive Man " :
" There can be no doubt that the free use of or privation
from metals is a question of life and death for any nation.
When we take into account the important part that is played
bv metals in all modern communities, we cannot fail to be con-
vinced that without metals civilization would have been im-
possible. That astonishing scientific and industrial movement
which this nineteenth century [1876] presents to us in its most
remarkable form, the material comfort which existing genera-
tions are enjoying, all our mechanical appliances, manufac-
tures of such diverse kinds, books and arts, — not one of all
these benefits for man, in the absence of metals, could ever have
come into existence. Without the help of metal, man would
have been. condemned to live in great discomfort; but, aided b}^
this irresistible lever, his powers have been increased a hundred-
fold, and man's empire has been gradually extended over the
whole of nature."
Origin of Iron Ores in Nature. — The popular scientific
theory is that in the beginning the matter which constitutes
the round globe upon which we live existed as a fiery mist floating
in space. In some way — we call it gravitation — the particles
of incandescent gaseous star dust (by mutual attraction) ran
together and condensed into a globular mass, which gradually
cooled, forming a solid crust, enveloped by dense, seething
metallic vapors, holding in suspension carbonic acid, oxide of
iron, siliceous sand, aluminous clay, magnesia, phosphates,
sulphate of lime, etc. At first, the metallic rain which poured
down boiled off again on approaching the heated surface. After
a time, however, this metallic rain ceased to rise again, and
remained part of the solidifying earth. Then came the birth of
vegetable life, colossal palms and immense ferns. It is well
30 2 The Iron and Steel Magazine
known that plants, under the action of light on their green cells,
decompose carbonic acid and liberate oxygen. Now the
earth's atmosphere at this stage — known as the carboniferous
period of geology, was densely laden with carbonic acid; but
in time this was abstracted and absorbed by the great forests
and swamps with their rank vegetable growths, thus clearing
the atmosphere and making the earth's surface habitable for
animal life. As the broad forests and widespread swamps be-
came submerged, either by the action of glaciers or subsidence
of the earth, the buried organic matter, consisting largely of
carbon, taken from the atmosphere, was petrified into coal;
while the immense areas exposed to the oxygenated atmosphere
gradually died, decomposed and decayed, evolving in the process,
carbonic acid, which attracted to itself the surrounding oxide-
of iron, silica, alumina, magnesia, sulphates, phosphates, etc.,,
and gradually solidified into beds of iron ore, even one hundred
feet thick, as in Lake Superior, U. S. A. (Fig. 3), and in India.
The various depths and angles at which these layers or strata
are found being mainly due to volcanic eruption and upheaval
of the earth's crust.
Having glanced at the nebular hypothesis of La Place, as
the best scientific explanation of the origin of iron ore in nature,
let us now follow this up by one or two illustrative proofs fur-
nished by modern research.
Aerolites. — Native iron in a metallic state is very rare.
It is one of the pet schemes of the metallurgical chemist to form
pure iron in the laboratory; but even then it has to be kept
hermetically sealed, otherwise it is attacked by the oxygen in
the atmosphere and quickly transformed into oxide of iron. If
exposed long enough it will crumble into dust. Nearly all iron
found in a metallic condition is of terrestrial origin, and is never
pure, for in the aerolitic form it is invariably alloyed with other
metals, such as nickel, cobalt, manganese, etc. The use of
aerolites in iron working is as old as history. Amerigo Vespucci,
after whom the North American continent is named, tells us that
in the fifteenth century the Indians at the mouth of the La
Plata River made their arrow heads of iron extracted from aero-
lites. Certain Siberian tribes are known to make their knives
from this source, and a like practice exists among the Laplanders.
Indeed, some writers have set up the fanciful theory that the
Dcsin'ptivc Mctalhtriiy of Iron ainl Steel
303
working of iron began with the use of these metallic stones
dropped from the skies. It is true these stones of terrestrial
origin are found in all parts of the earth, varying in size from
mere dust grains to masses weighing tons.
Fig. I is a photographic picture of " Ahnighito," the larg-
est meteoric mass of native iron ore known to be in existence in
the world. It was discovered by Commodore Robert E. Peary,
United States Navy, at Cape York, Greenland, in 1894, and is
now in the American Museum of Natural History, New York.
It is II feet long, 7^ feet thick, and weighs 37 tons. The
Reproduced by permission from " St. Xich(jlas Ma.orazine," March, 1905
Fig. I. Largest Aerolite Known to be in Existence
mass of metallic iron is alloyed with 8 per cent nickel and a
little cobalt.
In Fig. 2 is shown a diagrammatic section of the Egremont
mine on the northwest coast of England, illustrating the remark-
able manner in which strata of red hematite ore are sometimes
found imbedded in thick layers of limestone. As the waters,
heavily laden with vegetable matter in process of decompo-
sition and hence highly carbonized, flowed through the lime-
stone, it dissolved out large quantities of the rock, thus forming
great cavities and caverns. Now the Old Red Sandstone is rich
304
The Iron and Steel Magazine
in iron. In fact its reddish color is due to oxide of iron, just
as the red color of the blood is due to iron. When, therefore,
either by grinding glacier or raging torrent, denudation of the
sandstone rocks took place, the oxide in the finely divided par-
ticles of sand was dissolved by the vegetable acids into carbon-
ate of protoxide of iron. As this solution of iron floated along
in the drift it was acted upon by the oxygenated atmosphere,
iridescent films would appear on the surface of the flood, indi-
cating that the protoxide had been transformed into peroxide
of iron. These insoluble films of iron oxide, being of higher
.specific gravity than the rest of the drifting material, and
!^Eo one
Fig 2. Egremont Mine, England
becoming heavy masses by the addition of new particles, sank
to the bottom, filling up the cavities and deep fissures, and
gradually solidified into solid beds of red haematite iron ore, as
illustrated in Fig. 2.
To a like origin may be traced the formation of the famous
bed of hematite ore in the Chapin Mine, near Iron Mountain,
Mich., U. S. A. (Fig. 3).
This is the greatest deep mine deposit of ore being worked
in the world to-day, and was opened in 1880. It consists
of four lenticular deposits, 2,500 feet long, 130 feet wide, depth
unknown. The ore contains 63 per cent of metallic iron, 0.07
per cent phosphorus.
Descriptive Mctalhiri:^}' of Iron and Steel
305
Phos phonis . — Mention of the phosphoric contents of the
Chapin Mine ore opens the way for an explanation of the reason
why the extensive beds of carbonate iron ores which abound in
the Est on Hills of Yorkshire, England, and in the broad seams
running from New Jersey, through Pennsylvania, down to Ala-
bama in the United States, contain as high as 2.75 per cent of
phosphorus; while the 1,000,000,000 tons of hematite iron ores
buried in the Mesabi range and Lake Superior district, U. S. A.,
are practically free from this mortal enemy of the steelmaker.
Popularly stated, there are three periods in geological time:
(i) Silurian: age of invertebrates; (2) Devonian: age of
vertebrate fishes; (3) Carboniferous: age of coal plants, ver-
OB JASPEfi.
€Z exAY siA.re
-— ^uAmrziTg.
^B OOLONIITt
^m SAND
.zoo
.900
-400
-500
•600
Fig 3. The Chapin Mine, U. S. A.
tebrates, amphibians and reptiles. It is a well established
induction of science that the existence of phosphorus in iron
ores is due to the remains of decayed fishes and animals. It is
also a fact that the solid framework of the invertebrate animals
which existed in the Silurian period consisted of carbonate of
lime, whereas the bony structure of the vertebrates, which existed
in the later Devonian and Carboniferous periods, was made up
of phosphate of lime. Now the high phosphorous ores of the
Cleveland di.strict of England, and Pennsylvania, U. S. A., are
all found in beds of the Carboniferous age, in close proximity
to the great coal measures, when fishes, amphibians and reptiles
abounded; and it is from the decayed bones of these extinct
3o6 The Iron and Steel Magazine
vertebrates that the excessive phosphorus is derived. On the
other hand, the Lake Superior hematite ores are all found in
the levels of the Silurian age when vertebrates did not exist;
only shells of moUusks, corals, crinoids, trilobites and other in-
vertebrates, as already stated. The solid framework of these
creatures consisted not of phosphate, but carbonate of lime,
hence the comparative freedom of these iron ores from phos-
phorus.
Modern Ore Making. — If proof is needed of the foregoing
theory of the origin of iron ores, the reader can have actual
demonstration before his eyes in Canada to-day. About mid-
way between Montreal and Quebec in the valley of the St.
Maurice, where the river flows from the north into the St.
Lawrence, is Lac a la Tortue (Turtle Lake), a body of water
four miles long, by one and one-quarter miles in average width,
situated in the middle of a swampy morass. The environing land
consists largely of sand, doubtless carried down from the archaean
rocks in the vicinage, by the erosive and grinding action of
glaciers.
Standing on the western shore, the traveler gazes in imagi-
nation upon a primeval scene. Innumerable streams and
rivulets may be seen winding and percolating their way down
to the lake, through the sand rich in oxide of iron. These run-
ning waters are laden with the decaying vegetable matter which
grows rank in the marshy lands, carrying with it quantities of
the sand, saturated with iron oxide. The organic acids evolved
by the decomposition of the vegetable stuff dissolve the oxide
of iron, which is carried to the lake. But as it floats down, this
solution of protoxide of iron is acted upon by the atmospheric air,
oxidation takes place and a remarkable phenomenon is per-
ceived. Patches of iridescent film appear on the surface of
the lake, looking like petroleum with its rainbow colors, indi-
cating that the soluble protoxide has been transformed into in-
soluble sesquioxide of iron. The reason this peroxide film appears
in patches is due to concentrationary action; the particles
aggregate themselves into batches, which sink to the bottom
of the lake in the form of cakes ranging up to ten inches diam-
eter or more; hence the term " cake ore."
This brown hematite lake ore contains 70 per cent of
metallic iron, and seems inexhaustible; for with the decay of
o
On
3o8 The Iron and Steel Magazine
each year's vegetation, new supplies of iron from the sands are
deposited in the lake. These rich lake ores have been used in
the St. Maurice furnace at Radnor since 1752. In 1775 one of
the lessees of the Radnor furnace aided the American colonists
by casting shot and shell — made from the lake ores — to be
used against Quebec.
Lac a la Tortue and a neighboring lake are the only known
instances of the kind on the American continent.
Deep mine formations like those of the Chap in Mine, and
lake bottom deposits similar to that of Lac a la Tortue, are,
however, mineralogical curiosities when compared with the
magnificent surface deposits of the Mesabi range in Lake Supe-
rior country. We are almost bewildered with the marvellous
prodigality of nature when we look upon awe-inspiring scenes
like that pictured in Fig. 4. Gazing upon this stupendous work
of man, we are impressed with the same sense of grandeur and
majesty as we are upon first beholding Niagara Falls; one
caused by the mighty hydraulic forces of nature, the other by
the engineering skill and inventive genius of man.
And this is only one of the series of enormous holes made
in scooping out with giant shovels 79,000,000 tons of ore in the
past twelve years. There is 75 miles run of these beds of soft
hematite ores, containing 50 per cent of metallic iron and
practically free from phosphorus.
Having described the modus operandi in which ores are
formed in nature, we now pass on to a consideration of the
properties of the ores of commerce.
{To he continued)
Metallography Applied to Foundry Practice
309-
METALLOGRAPHY APPLIED TO FOUNDRY PRACTICE *
PART III
By ALBERT SAUVEUR
Microscopical Examination of Prepared Samples
\ FTER a sample of cast iron has been polished and etched
'^^ as described in the first two installments of this article, it
is ready to be examined under the microscope. The apparatus
needed to carry on this examination will be briefly described.
The Microscope. — While any good microscope of the or-
dinary tvpe may be used with some degree of success for the
examination of samples of metal, a special stand, as illustrated
in Fig. I, is almost a necessity for effective work. It differs from
the ordinary type by the motion
of the stage, which can be raised
or lowered by rack and pinion
mechanism. The advantage of
this feature will be appreciated
later on upon considering the
illumination of the sample.
The microscope shown in
Fig. I is manufactured by the
Bausch & Lomb Optical Com-
pany, of Rochester, N. Y., and is
the instrument universally used
in the United States for the ex-
amination of metals. The micro-
scopical outfit should include
one low-power objective (one half
or two thirds of an inch focal
length) , one high power objective
(one sixth or one eighth of an
inch focal length) and two eye-
pieces, respectively of one-inch
and two-inch focal length. These
will yield the different magnifi-
cations and resolutions which xke ro..narj
are needed for a successful examination of the structure of cast
* " The Foundry," September, 1905.
3IO The Iron and Steel Magazine
iron, it being very seldom that higher or lower magnifications
are found necessary.
When examining a specimen of metal under the microscope
it is absolutely necessary to bring the polished surface in a plane
exactly perpendicular to the axis of the instrument, and unless
some special device be used for holding the sample in this posi-
tion, this means that the polished surface and the opposite
side must be exactly parallel, so that when the sample rests upon
the stage of the microscope, the upper surface will be in the
proper position. The difficulty and labor implied by the cutting
of two exactly parallel surfaces will be readily appreciated. To
obviate this necessity, the special specimen holder illustrated in
Fig. 2 will be found most effective. A rubber band holds the
Fhj. 2
The I'uuuihy
Specimen firmly in the holder, however irregular in shape, and a
cover glass may be inserted between it and the object so as to
fulfill the requirements called for by the correction of the ob-
jectives. The holder is then placed on the stage where it can
be clipped like any ordinary microscopic slide. (See Fig. i.)
Illumination of the Sample. — Metallic samples are absolutely
opaque objects and, unlike the ordinary transparent slides of
other microscopists, cannot be illuminated by transmitted light;
that is, by light sent from below the stage through the object.
Reflected light, that is, light thrown from above upon the
specimen, must of necessity be used. This light may be reflected
obliquely upon the specimen, directly from the source of light or
with the assistance of a mirror or of a condenser, or it may be
AIciaIIoi:^rapliy Applied to Foundry Practice
311
reflected vertically upon the sample by the lenses themselves of
the objective by means of a device known as a vertical illumi-
nator and which affords by far the best illumination of opaque
objects. This illuminator, which is shown in Fig. 3, consists
of a small glass disk, supported by a milled head, which controls
its position, and is inserted between the objective and the nose-
of the microscope, as shown in Fig. i. The light from the
source of light enters the illuminator through a lateral aperture
and is reflected downwards by the little glass disk through the
lenses of the objective, which condenses the light upon the
Figs. 4 and 5
sample. The light emitted by the lighted portion of the speci-
men re-enters the objective, passes in part through the glass
disk, then through the eye-piece, reaching the eye of the observer
and producing an enlarged image in the usual way.
Obliquely vs. Vertically Reflected Light. — It will readily be
seen that obliquely reflected light can be used only with rela-
tively low power objectives, because high-power lenses must be
brought too close to the object to allow any beam of light to
reach the latter from without. Obliquely reflected light, more-
over, yields, so to speak, a negative image of the true appearance
of the sample, because the bright constituents reflecting the
312
The Iron and Steel Magazine
light outside the microscope will appear dark, while the dark
constituents, by absorbing and diffusing the light, reflect some
of it into the tube of the microscope and appear bright. Ver-
tical illumination, on the contrary, yields a true image of the
appearance of the object.
In Figs 4 and 5 are shown the structure of exactly the same
spot of a sample of gray cast iron, but illuminated respectively
Fig. 6. Special Arc Lamp for Opaque Objects
with obliquely and vertically reflected light. It illustrates
strikingly the opposite effects produced by these two kinds of
illuminations. In Fig. 4 the plates of graphite appear bright,
while they are black in Fig. 5, and therefore with their true
color. The iron matrix, on the contrary, is dark in Fig. 4 and
light in Fig. 5. In view of these considerations, it will be ob-
vious that vertically reflected light is generally the best means
of illuminating metallic samples.
Metallography Applied to Foundry Practice
S'^S
Sources of Liglit. — Many artificial lights may be used to
obtain a satisfactory illumination of metallic samples, but it will
suffice to mention here the two sources which are most generally
used, namely, the electric arc lamp and the Welsbach lamp.
Fig. 6 illustrates a special arc lamp and accessories for the illumi-
nation of opaque objects. The outfit consists of an electric
Fia. 7
' Tic: F^uuiiy
90-degree arc lamp and rheostat, a triple condenser system, a
cooling cell, an iris diaphragm and automatic shutter, an optical
bench and base board. The illustration also shows the micro-
scope connected with a camera for taking photo-micrographs.
In Fig. 7 is seen a simple Welsbach lam.p and condenser, an
arrangement which ^aelds a very satisfactory illumination for
examination by low-power objectives.
314 The Iron and Steel Magazine
SPECIAL STEELS *
By L. GUILLET
'"T^ERNARY alloys and steels are those in which the alloy
■■' consists of iron, carbon and an intentionally added metal,
the other constituents being kept within the usual limits. In
the following description of certain of these steels, the author's
investigations have been concerned with two grades, one con-
taining only about o. 2 per cent of carbon, the other approxi-
mating to the eutectic and containing about 0.8 per cent of
carbon.
The mechanical tests were of three classes, — tensile strain,
shock by the Fremont method and hardness by the Brinell
method. The tensile tests were performed on standard test
bars, but with the tungsten steel the bars were 200 mm. long
between the punch marks, the others measuring half that
length; the diameter in all cases was 13.6 mm.
Nickel Steel. — This steel has received so much attention
in the "Journal of the Iron and Steel Institute," and in the
papers of Riley, Osmond, Dumas, Hadfield, Hopkinson, Guil-
laume and others, that we do not propose to review at length
the author's investigations on this alloy.
Manganese Steels and Chrome Steels. — The remark made
above applies equally to these alloys.
Tungsten Steel. — There is no difficulty in producing tung-
sten steel in the crucible furnace with a high temperature. The
usual proportions of tungsten are 0.5 to 13 per cent (except for
spring steel, which will be referred to later). Either ferro-
tungsten or the metal itself may be employed; the former,
containing 8 to 9 per cent of carbon and about 80 per cent of
tungsten, is prepared in the electric furnace, while the metal
is obtained either by this means or the aluminothermic process.
The diagrams furnished by the mechanical tests performed
on the crude steel demonstrate that the, breaking strain of the
pearlitic steels increases in proportion with the tungsten content.
The limit of elasticity does not increase with the same rapidity,
while the reduction of area and elongation suffer little diminu-
* From a paper presented before the Mining and Metallurgical Con-
gress in connection with the exhibition at Liege, Belgium, June, 1905.
special Steels
315
tion. Under the Fremont test their fragility is not greater than
that of carbon steel, and as a rule their hardness is superior to
that of ordinary steel with the same carbon content. The car-
bide steels have a high breaking strain, directly proportional
with the percentage of carbon, but not increasing in proportion
to the amount of tungsten. The limits of elasticity are rela-
tively low and the reduction of area and elongation very small.
These steels border on the class of brittle steels and present the
remarkable feature that their resistance to shock is nearly con-
stant, whatever the percentages of carbon and tungsten present.
Finally, the hardness increases with the proportion of carbon.
1 5 1.6
CAPBON %
Fig. I. Diagram of Tungsten Steels
If a diagram be constructed of the tungsten steels, Fig. i,
the straight line drawn from the point indicating 1.6 per cent
of carbon to that relating to 10 per cent of tungsten will divide
the steels into the groups of pearlitic and carbide steels re-
spectively.
At present the tungsten steels are chiefly used in the manu-
facture of tools, the low-grade steels requiring to be tempered
in water, while the high-grade varieties are self-tempering,
though they require an admixture of chromium or manganese
for that purpose. Some steel makers have put tungsten spring
steel on the market, the average composition of the product
3i6
The Iron and Steel Magazine
being: Carbon, 0.47; manganese, 0.22; silica, 0.20; and tung-
sten, 0.6 per cent.
The untempered steels give the following values: breaking
strain, 113,500 to 120,900 pounds per square inch; elongation,
14 per cent; elasticity, 85,000 pounds; these being modified
by tempering and suitable annealing to 199,000 pounds, 7 per
cent, and 142,000 pounds, respectively. These steels do not
seem to be much superior to the silicon steels generally used,
being very high in price and more brittle than silicon steel when
tempered and annealed.
Molybdenum Steels. — These are prepared in a manner
similar to that employed for tungsten steel. Under mechanical
tests the pearlite steels (0.5-2 per cent of Mo) have a far higher
breaking strain than ordinary steels, the elongation and reduc-
1.5 1.6
CARSON <i
Fig. 2. Diasfram of Molvbdenum Steels
lion of area being also very good. They are, moreover, harder,
though not more brittle, than carbon steels. The double car-
bide steels have very high breaking strains and limits of elas-
ticity, with only slight elongation and reduction of area, and
they are brittle and extremely hard. The line of demarcation
between the groups of pearlite and double carbide steels runs
from 1.6 per cent of carbon to 2.5 per cent of molybdenum
(Fig. 2).
Molybdenum steel was put upon the market as a special
tool steel of some mysterious quality, but has since been aban-
doned, the steel being not much better than tungsten steel,
while the price is higher, though the proportion of added metal
is smaller. Nevertheless, some makers continue to use molyb-
denum and tungsten together.
special Steels
317
Vanadium Steels. — The tests above mentioned were ap-
plied to vanadium steels that had been slightly annealed by
heating to 900° C. and slowly recooling, this treatment being
rendered necessary by the fact that these steels are more pro-
foundly modified by working than others. The grades high in
vanadium are very irregular, owing to the low density of the
double carbide leading to its uneven distribution throughout
the mass. Tests applied to different parts of the same bar gave
the following results at the two ends:
Steel with 0.382 per cent of carbon and 5.37 per cent of
vanadium: breaking strain, 48,000 to 78,000 pounds; elas-
ticity, 22,000 to 49,000 pounds; elongation, 23 to 17 per cent.
< 3
<-' 0.8 CARBON %
Fig. 3. Diagram of Vanadium Steels
Steel with 0.130 per cent of carbon and 7.37 per cent of
A^anadium: breaking strain, 40,000 to 75,000 pounds; elasticity,
28,400 pounds; elongation, 21 to 17 per cent.
Steel with 0.120 per cent of carbon and 10.27 per cent of
vanadium: breaking strain, 43,000 to 77,800 pounds; elasticity,
29,800 to 67,000 p)ounds; elongation, 22 to 21.5 per cent.
Steel with 0.737 P^^ cent of carbon and 7.85 per cent of
vanadium: breaking strain, 43,000 to 77,800 pounds; elasticity,
19,000 to 57,000 pounds; elongation, 16 to 22 per cent.
Steels with 0.858 per cent of carbon and 10.25 P^^ cent of
vanadium: breaking strain, 59,000 to 100,200 pounds; elas-
ticity, 24,100 pounds to 44,800 pounds; elongation, 10-10 per
cent. Here the breaking strain and elasticity increased pro-
3i8 Tlie Iron and Steel Magazine
gressively from one end of the bars to the other, the increase
corresponding to higher percentages of carbide (determined b}"
the micrographical examination of the fractured surfaces).
The diagram furnished by these steels shows that in addi-
tion to the pearlite and double carbide groups, they form a third
group containing both these components; also that the larger
the percentage of carbon, the more vanadium is required to
keep the whole of the carbon in the condition of carbide (Fig. 3).
The net result of the tests is to exclude from industrial
application all grades containing over 0.7 percent of vanadium,
such of them as contain carbide being practically useless. The
characteristic property of vanadium is to impart hardness to
the steel and increase the breaking strain and limit of elasticity.
In fact, in this respect, the metal plays a part similar to carbon
and at least as powerfully as the latter. Another interesting
property of vanadium is that in raising the breaking strain and
limit of elasticity, it does not diminish the elongation, nor
does it produce the slightest fragility; these remarks apply-
ing, of course, to the permissible limit of the metal as given
above.
It follows that nothing stands in the way of an extensive
utilization of vanadium steels (which are not difficult to manu-
facture), provided they can be produced cheaply. At present,
the cost of ferro-vanadium is failing, for whereas a short time-
ago it was worth about 445. per pound of vanadium present,
American makers have latterly offered it as low as 145. 6d.
It is evident, therefore, that since the most valuable vanadium
steels are those containing 0.2 to 0.5 per cent, the extra cost
of adding these quantities to ordinary steel is really negligible
(65. to ids. per ton). Nevertheless, it must not be forgotten
that vanadium steels are very sensitive to thermal and mechani-
cal treatment, and that they cannot be used until they have-
been carefully annealed by heating to 900° C. and gradually
recooled. The author is convinced that steels containing less
than 0.7 per cent of vanadium have a great future before them,,
comparable in every Way to that of pearlite nickel steels.
Titanium Steels. — Much has been heard of the advantages
of titanium steel in a general way, without any definite par-
ticulars respecting their valuable properties; and for this;
reason a systematic study of the series appeared to be of in-
special Steels 319
terest. The specimens experimented upon were prepared with
ferro-titanium produced in the electric furnace, but could not
be made of higher titanium content than 10 to 15 per cent, the
temperature of the ordinary crucible furnace being insufficient
to fuse the higher alloys. The results were, however, disap-
pointing, so far as any improvement in the breaking strength,
hardness and resistance to shock are concerned, these steels
being devoid of any practical interest. This circumstance does
not preclude the possibility of using titanium as a purifying
agent.
Cobalt Steels. — There is no difficulty in the manufacture of
these steels, metallic cobalt being used. Its influence on the
properties of the iron differs from that of nickel, at least up to
a cobalt content of 30 per cent and a carbon content of 0.8 per
cent, the products being all of the pearlitic group.
The mechanical tests, too, show that apart from a slight
increase in the breaking strength, cobalt has very little influence
in any proportion between 4^ per cent and 30 per cent; con-
sequently, the cobalt steels are of no industrial interest.
Tin Steels. — Tin acts much in the same way as silicon and
titanium, but does not play the same important part as the
former in steel; it will not precipitate carbon in the form of
graphite. Up to nearly 5 per cent it is dissolved by the iron,
but at that Hmit a definite stannide of iron seems to be formed.
The carbide, up to 10 per cent, at least, always remains in the
form of pearlite.
Steel containing 1.5 per cent of tin is very difficult to forge,
and above 2 per cent renders it unworkable. The steel becomes
extremely brittle, breaking into small fragments under the
shock of conveyance from place to place; and if the bars are
dropped from a height of about three feet on to a flagstone, they
are reduced almost to powder. Though a sufficient percentage
(about 5 per cent) of tin makes the steel very hard, there is no
practical use for any member of the series.
Classification of Ternary Steels
The special steels may be classified in the following manner:
(i) The added metal lowers the conversion points of the iron,
and gives three groups, — pearlite, martensite and ;--iron steels.
'This class comprises nickel and manganese.
320 The Iron and Steel Magazine
(2) The con Aversion points are lowered and, in addition, a
double carbide is formed, the three groups then being, pearlite,
martensite and double-carbide steels. Chromium belongs to this
class.
(3) The added metal (tungsten, molybdenum and vana-
dium) is capable of forming a carbide, so that we have two
groups: pearlite and double-carbide steels.
(4) The added substance precipitates carbon in the condi-
tion of graphite, and gives two groups of steel, one containing
combined carbon, the other graphitic. This is the case with
silicon.
(5) The metal dissolves in the iron, and gives in all cases
pearlite steels. Such metals are cobalt, titanium and tin.
The micrographic character of special steels furnishes in
some instances vahiable indications as to their mechanical prop-
erties. In the case of pearlite steels, however, these indica-
tions are merely partial, the mechanical properties depending
on the element dissolved in the iron or in the pearlite (for
instance, a pearlite tin steel is of no use). Hence, mechanical
tests must be made to determine the quality of the steel. With
a martensitic steel, on the contrary, it may be at once concluded
that it has high breaking strength and limit of elasticity, with
low elongation and slight reduction of area, is brittle, hard and
difficult to work. A polyhedral steel will have a rather low limit
of elasticity but high elongation, great power of resisting concus-
sion, and may be more or less hard according to the substances
dissolved in the pig iron. If the structure of the steel indicates
a carbide, chrome steel is revealed by round grains, tungsten, or
molybdenum steel by fine filaments, and vanadium steel by
large triangular granules. The presence of graphite in steel
argues extreme fragility and practically no elongation or reduc-
tion of area.
Uses of Ternary Steels. — A certain number of pearlitic
steels can and should be used in place of carbon steels. This
is principally the case with nickel steels, and to some extent also
with manganese steels.
Of the martensitic steels, a few only are suitable for con-
structional uses, owing to their general fragility and difficulty
in working them. Some, however, may be used in making tools
or for particular purposes.
Hot Cracks in Steel Castings 321
The polyhedral steels, if cheaper, would have a great future
before them. At present they can only be used in special cases,
and, moreover, they have a very low limit of elasticity.
Among the double-carbide group, the tvmgsten and molyb-
denum steels are likely to be in good demand, by reason of their
great hardness on tempering. Besides, they are already in
general use as tool steel. Finally, the graphite steels are useless
for practical purposes.
The author also deals with some quarternary steels, i.e.,
those in which the steel is alloyed with carbon and two other
intentionally added elements. The steels dealt with are the
nickel-chrome, chrome-tungsten and manganese-silicon steels,
which are in every-day use. Experiments have also been made
with nickel-manganese, nickel-silicon and nickel-vanadium steels,
which have not yet come into general use. The consideration
of these must, however, be deferred.
HOT CRACKS IN STEEL CASTINGS*
By ARTHUR SIMONSON
A X attempt will be made in the following article to outline
'^^ the principal causes of one of the greatest difficulties that a
steel founder has to contend with, and to suggest some means by
which it may be overcome, at least partially. Cracks in steel
castings are of two kinds, which differ in their appearance and
cause very materially. Hot cracks take place at the time of
solidification of the metal or very soon after; cold cracks are
formed while the metal is below red heat. The former take the
appearance of a tear, are very ragged and there is a sinking of
the metal at the edges; they are generally quite wide and have
a film of blue or black oxide on their fractured surfaces. Cold
cracks, while they may be open occasionally, are generally very
fine, clean cut as with a knife, and unless the castings are care-
fully inspected may sometimes escape observation. Ringing
the castings with a hammer will often reveal the presence of cold
cracks which are almost invisible. It is with the former, or hot
cracks, that the present article is intended to deal.
* " The Foundry," August, 1905.
322 The Iron and Steel Magazine
The two principal causes of hot cracks are obstructions ta
the free contraction of the metal, and unsuitable composition of
the metal itself. First, then, look into the causes which may
prevent the unrestricted contraction of the metal. They are
chiefly the rigidity of the mold and the varying thicknesses of
section of the casting. The mold has to be made sufficiently
strong to stand the weight of the steel and the fluid pressure of
the head of metal while it is being poured. Molds for steel cast-
ings are generally made in dried sand, which consists of silica
sand mixed with a certain proportion of clay to bind it together.
And though it is very weak in its green or damp condition, it
becomes quite hard and firm after baking. The molds are faced
with a wash made of silica flour and molasses water, which gives
a very hard, refractory skin. It is, therefore, important that
while the mold should be strong enough to stand all the pressure
it is to receive, it should not be any stronger than is necessary
for the above purpose. Means may be provided for making the
mold stronger in some parts than others, for instance near
the gate, where the cutting action is greatest. At these places
the mold may be made of a stronger grade of sand, or, if its
shape allows, hard cores or firebricks cut to shape may be fitted
in, to take the wear of the stream of metal. All square corners,
both inside and outside, should be amply filleted, and wherever
a rib or a projecting arm of the pattern protrudes, the sand in its
immediate vicinity should be loosened up by ramming in cinders,
sharp sand or sawdust; or the mold can be cut away to within
two or three inches of the pattern, after it has been dried, and
the space filled in with burnt sand.
Another point to be attended to with the idea of reducing
the danger of hot cracks is the drying of the molds. To produce
the best results a mold should be rather over-dried than under-
dried ; that is to say, it should be almost but not quite burned. A
mold that is only just dry is in the most rigid possible condition ;
it can be baked a good deal more and yet preserve sufficient
strength to stand the wear and tear of pouring, and it will then
offer much less resistance to the shrinkage of the metal. The
ideal mold, as has been said before, consists of a hard refractory
skin and a collapsible backing, which will give way as soon as
the cooling skin of the casting has become sufficiently rigid to
support itself, and begins to shrink. It is to the production of
Hot Cracks in Steel Casiiiigs 325
these conditions as nearly as may be possible in practice that
foundrymen have to bend their efforts.
Defective construction of cores is another fruitful source of
cracked castings. Coremaking is a branch of the steel foundry
trade that does not receive the attention it merits. It is equally
as important as the mold itself, calls for as much skill and con-
tributes equally to the success or failure. And 3^et we often find
the coremaking relegated to a very secondary place. Core sand
mixtures should be as carefully studied as molding sand mix-
tures, and a great saving may be effected, not only in the matter
of cracking, but in the cost of cleaning and the soundness of the
castings by careful attention to this point. The same descrip-
tion applies to a core as to a mold,- — it should have a hard,
smooth face, which will resist the cutting and fusing action of the
metal, but it should crumble and fall out in the form of powder
when burnt. Careful handling will permit the use of cores which
seemingly are exceedingly delicate. As the writer has previously
stated, cores can be made of almost anything, provided the wash
is all right. When the core is rammed up it should have a good
coat of a wash made of silica flour, ceylon graphite and molasses
water, and then put in the oven and baked until after scratching
the skin the inside is thoroughly " rotten." Then another coat
of wash, or two if necessary, may be given, and the core is redried.
It is surprising how strong this skin becomes, and it is no more
than one thirty-second of an inch thick.
In a great many cases a core has to stand much greater
pressure than the mold itself, as, for instance, in a pipe or cylinder,
where the metal is shrinking on the core from every direction.
If the core is not collapsible one of two things must happen, —
either it will crack the casting or the core will become so hard
that its removal will be a very expensive and lengthy operation.
The second point, namely, the composition of the metal itself,
is equally important with the foregoing. Any conditions which
tend to hot shortness of the metal, which means brittleness above
red heat, must be carefully avoided. The two principal ele-
ments found in common practice which have this tendency are
sulphur and copper, and while their influence is not very great
in the cold state of the steel, still, as the metal has to pass through
the hot short period before cooling to the ordinary temperatures,
it is important they should be kept as low as possible. Either
324 The Iron and Steel Magazine
of them by itself is dangerous, but the combination of the two
is fatal. As a large proportion of eastern iron is made from
ores from the Cornwall district, a great deal of the scrap avail:
able, as well as the iron, has an appreciable content of copper,
and it is therefore necessary to watch the sulphur most care-
fully, and care should be taken not to allow it to run over .045
per cent. This is done by selecting melting stock as low as possi-
ble and running a high manganese, which will prevent increase
of sulphur from the coke, etc., and tend to reduce it, if anything,
by the formation of sulphide of manganese.
COST OF PRODUCING STEEL CASTINGS BY THE OPEN-
HEARTH PROCESS AND THE SMALL CONVERTER *
By L. UNCKENBOLT
'"T^HE small open-hearth furnace and the small converter
are not really rivals, each having its proper field of action
and suitable uses, and any attempt at replacing one by the
other will prove an economic failure. A four- to five-ton open-
hearth furnace for cast steel can only produce castings exceed-
ing about half an inch in thickness and i cwt. in weight,
without wasting three to four tons of the charge, the contents
of the furnace cooling down so quickly during the casting that
the steel becomes insufficiently fluid to run into small molds.
Now the converter enables castings of any size to be made, and
is particularly profitable when those of small dimensions (less
than one-half inch thick and i cwt. in weight) are to be pro-
duced; in fact, Belgian foundries make, by its aid, castings of
one pound weight and one-eighth inch thick, such as could only
be obtained formerly with crucible steel, malleable castings,
or the hydraulic press. However, it is not advisable to erect
a steel plant solely for the production of small goods on which
a large profit is made, the work of casting being in this case very
complicated. In order to avoid this inconvenience, it is neces-
sary to produce large articles as well, in a proportion that can
easily be determined by experience.
The cost of a simple steel plant, with a single cupola capa-
ble of treating about three tons per hour, and a one-ton con-
* " Iron Trade Review," August 17, 1905.
Cost of Producing Steel Castings 325
verter, will be, for the cupola, £^o\ the fan, ;^2o; the converter,
£200; the blower, ;^-ioo; the reheating furnace, £^0; the drier,
£^o\ the sand blast machine, ;£i2o; two drums and two mills,
£'40; several casting ladles and other accessories, ;)^ 120; crusher
and disintegrator, for sand ,-£40; small pattern shop, £80; total,
£880.
Should the business after a while entail an increase in the
plant, the following additions could be made: a second drier,
£60; a second reheating furnace, £60; an electric welder for
repairing small superficial defects, £200; several grinding ma-
chines, £100; one or two cold saws, £60 \ a small chemical
laboratory, £120; a testing bench, ;^8o; total, £680.
To this outlay must be added the cost of buildings, source
of motive power, a hoist, a platform for the cupola, a weigh-
bridge for the yard, and a five-ton traveling foundry crane, the
whole amounting to £2,000 to £3,000.
Statements have recently appeared in the German tech-
nical press to the effect that small converter steel plants can be
erected at less than the above figures; but it is hardly probable
that such plants would be able to compete with those already
existing, it being essential to success to furnish consumers with
a good article, quickly and at a low price.
The motive power required will be 5 to 10 horse-power for
the fan, 5 horse-power for the hoist, 50 horse-power for the con-
verter blower, 5 to 10 horse-power for the sand-blast machine,
10 to 15 horse-power for the drums, 5 to 10 horse-power for the
sand-preparing machinery, and 3 to 5 horse-power for the pat-
tern shop. Later on, this can be increased by 5 to 10 horse-
power for working the converter, 30 to 40 horse-power for the
electric welder and 10 to 15 horse-power for the saws, lathes,
traveling cranes, planing machines and testing bench. It will
not be necessary to keep all the machines working continuously.
For example, to produce one ton of steel, the cupola fan is worked
for half an hour, the converter blast half an hour, the converter
mechanism three to four minutes, and the hoist five minutes.
In this way, a 100 to 500 horse-power engine will be sufficient,
though it is always better to have too much power than too little.
A traveling crane of 5 to 7 J tons capacity is a prime essential,
since it will enable all orders to be executed, even with only a
single cupola and one converter.
326 The Iron and Steel Magazine
The steel from the converter is at such a high temperature
that several charges can be collected into a large heated ladle,
so that very large castings can be produced when required.
The author has made castings of 3 to 5 tons each in this way
without difficulty.
The cost price may be calculated as follows: To obtain one
ton of molten steel, as it issues from the converter, the materials
required will be: 14 cwt. of Bessemer iron, at 565. = 396". 2d.;
4 cwt. of hematite pig, at 625. dd. = 12s. 6d.; 5I cwt. of scrap,
at 44s. = lis. 6d.; | cwt. of complementary materials (ferro-
manganese, ferro-silicon, ferro-aluminum) = 55. Sd.; 4% cwt.
of foundry coke, at 175. 6d. = 4s. 2d.; 2-| cwt. of gas coke for
heating the ladles, drying the molds, etc., = 15. 6d.; or, together,
745. 6d., as the cost of the ton of steel. From this quantity
must be deducted 5 cwt. for waste in casting, and worth only the
price of scrap (44s.), viz., 115., leaving 15 cwt. of steel costing
635. 6d., i.e., 845. 8 J. the ton.
The cost has to be increased by the items for labor, upkeep
and motive power, as follows: In respect of labor, there will be
needed for an output of 25 tons, two men for the cupola, at 35.
6d. = 75., and two for the converter at the same rate of pay,
or 145. in all, which works out at a little over "jd. per ton. With
regard to the upkeep of the cupola and converter, a i-ton con-
verter working every day will usually give 60-80 pourings before
needing repair. But assuming that it needs repair once a fort-
night, and that during this period 50 pourings have been made,
the refractory lining will cost £$ and the labor 145., which,
together with 6s. for the repair of the cupola, makes ;^6 for an
output of 50 tons, or 2s. ^d. per ton.
To produce 25 tons a week will entail the following outlay
for motive power: 5 horse-power for i\ hours to work the hoist,
costs i\d.; 10 horse-power for 12^ hours driving the cupola fan,
25. and 50 horse-power for an equal period, driving the con-
verter blower, costs 10s. or 125. i\d. in all, making in round
figures td. per ton. The total cost of the ton of steel is, there-
fore, 845. 8(i -f 7(i. -f- 2S. ^d. -{■ 6d. -=•■ 885. 2d. This may seem
high, but is an outside figure, having been obtained on the basis
of an output of 20 to 25 tons per week with a plant consisting
merely of one cupola and a single converter. Hence the cost
would be considerably reduced in a plant with two converters.
New Open -Hearth Steel Process 327
since in the latter case a larger number of pourings could be
made without interruption, thus greatly diminishing the outlay
for coal and coke.
No doubt the objection may be urged that steel can be pro-
duced much cheaper in the open-hearth furnace, but it should
be remembered that this furnace, when once started, must be
kept going at all hazards, whatever the state of the market, a
condition that will undoubtedly prove onerous in the long run.
With the converter, on the other hand, work can be suspended
when trade is bad and orders slack, so that no loss is being
incurred, even if no profit is being madcr
NEW OPEN-HEARTH STEEL PROCESS *
By P. ACKERS
OINCE the year 1865, the date of its appearance, the open-
hearth process of steel manufacture has made enormous
progress. From the 5-ton furnace built by Martin at Sireuil,
to the 200-ton furnaces recently erected by Wellm.an in America
and in England, there is a great step, almost unique in the annals
of metallurgy. The Martin process owed its initial successes
to the use of scrap and crop-ends from Bessemer steel works
and mills; this success grew greater when it became recognized
that the metal obtained in the open-hearth furnace was more
regular and more definite in its composition than that obtained
by blowing processes. From thenceforward Martin steel was
employed in all special manufactures requiring a metal of specific
chemical composition and manifesting definite mechanical
properties. The j^rocess, as usually carried out, consists of
charging a certain quantity of solid pig iron, and of completing
the charge with scrap, the proportion of pig iron varying accord-
ing to the heat of the furnace, the quality of the iron and the
degree of hardness required in the steel.
This process, which x>ermits of the requisite degree of puri-
fication being attained even with phosphoric pig iron and in-
ferior scrap, takes a longer time the more phosphorus there is
in the bath. Even with a pure pig iron requiring but a slight
* Read at the Mining and Metallurgical Congress at Li^ge.
328 The Iron and Steel Magazine
degree of dephosphorization the production cannot be reckoned
on exceeding 70 tons per 24 hours in a furnace of 1 5 tons capacity.
This being so the cost of open-hearth steel is far higher than that
of Bessemer or basic steel, but the advantage of the process
lies in the fact that, provided the sulphur be low, a wide range
of pig iron can be employed, whereas the other processes necessi-
tate the use of pig iron containing definite amounts of silicon
and of phosphorus, which prevents the use of many ores.
Nevertheless, the manufacture of steel by the open-hearth process
is slow, and only yields steel at more or less lengthy intervals
and in large quantities at a time, which is a drawback from the
point of view of securing a regular supply for the mills.
All new processes of open-hearth steel have as their object
the avoidance of these disadvantages and the regularization ,
as far as possible, of the furnace outputs. One improvement
that has been effected is the use of liquid pig iron, introduced
straight from the blast fu.rnace, into the furnace. This method
is, however, only logically sound when a large proportion of
the charge, say 75 per cent, is introduced in this way, under
which conditions the decarburization of the bath will take even
longer than usual unless it is hastented by the addition of some
material which is m^ore highly oxidizing than scrap, i.e., iron
ores.
Ore Process. — The method emplo3ed in this case is based
on the old direct ore process, of which it is but a perfected out-
come. It consists of charging the ore and the liquid pig iron
together. The ore is charged first and heated in the furnace
until the ladles of liquid pig are brought up and their contents
poured into the furnace. The decarburization of the pig iron
in contact with the mineral oxides takes place more rapidly than
in the old process, and produces a violent reaction which hastens
the end of the operation; in five hours from the time of charg-
ing, 25 to 30 tons of steel may be run out. The ores emiployed
should be rich and easily reducible, and they should be in suit-
able-sized pieces, the presence of much fine ore occasioning
the passage of much dust into the regenerators. The pig iron
should have a suitable proportion of manganese; the presence
of phosphorus, while delaying the operation, scarcely affects
the ultimate result. The process also economizes a consider-
able quantity of fuel. The jnethod was first practiced at Dona-
Neiv Open-Hearth Steel Process 329
witz, and has spread rapidly since, being peculiarly adapted
to districts where the available ores are too high in phosphorus
for the Bessemer process, yet too low for the basic process.
Silesia and Poland afford instances.
The following details relating to the practice followed at
the Jurjewka Works, Donetz, were furnished by Mr. Stmdgsen,
the managing director.
There are five 25- to 30-ton furnaces supplied with metal
from a 150-ton mixer. The metal shows the following average
composition :
Silicon Manganese Sulphur Phosphorus
i.otoi.5 2.25103.0 o. to 0.3 0.15100.25
Krivoi-Rog ore containing 65 pet cent of iron and o.io per cent
of phosphorus is used.
The results obtained at these works are remarkable. Dur-
ing July, 1904, 13,754 tons of good ingots were made on a work-
ing equal to 140 da3^s, and 570 charges, or 98 tons per 24 hours
per furnace. A 20 per cent saving of coal was effected. The
average composition of a charge was 25,724 kg. of liquid iron and
176 kg. of f erro-manganese ; 6,240 kg. of ore and 2,400 kg. of
fluxes. The working of the charge occupied 5 J hours from
-charging to tapping. The composition of the pig iron used
.and of the steel obtained were as follows:
C
Si
Mn
S
Ph
'ig iron . .
• •• •4-25
1. 17
2.80
0.03
0.20
teel
. . . .0.07
O.OI
0.50
003
0.04
Five thousand kg. of slag were produced containing:
Si Mn S Ph Fe CaO AI2O3 MgO
24.8 14.40 0.14 0.46 8.10 35-55 i.8o 9.7
This corresponds to a loss of iron equal to 405 kg. The ore,
"however, yielded 4,056 kg. of iron, so that the gross gain of iron
equaled 3,651 kg. The loss on metalloids, manganese, etc.,
amounted to 2,020 kg.; hence, 3,651 minus 2,020 represents the
actual yield over the weight of pig iron charged = 1,631 kg.,
which is equivalent to a yield of 106 [jer cent.
The furnaces at Jurjewka resemble in all respects ordinary
basic open-hearth furnaces of 25 to 30 tons. When the liquid
pig iron is first charged into the furnace on to the ore, there is a
-very violent reaction. When the boil becomes too strong, the
330 The Iron and Steel Magazine
air and gas are temporarily cut off and other precautions are
taken. It is necessary that the gas ports should be fairly long
in such furnaces, and that the angle at which they enter the
furnace should be sufficient to prevent slag finding its way into
the regenerators when the reaction becomes too intense. At
Jurjew^ka the angle is about 30°, and the length nearly to ft.
While the process effects various economies in cost as compared
with that usually employed, it has one disadvantage, — the
difficulty of regulating the supply of pig iron from the blast
furnaces to the mixer, arising at times from the simultaneous
tapping of two blast furnaces, or the simultaneous charging of
two or more steel furnaces.
The Bertrand-Thiel process of steel prodviction entails two
distinct operations and the employment of two furnaces or sets
of furnaces. It depends on the following considerations. If
liquid pig iron is brought into contact with less ore than is re-
quired to oxidize its impurities a rapid reaction occurs, and
the pig iron exhausts the oxidizing capacity of the ore, with the.
production of an inactive slag and the reduction of a large pro-
portion of the iron in the ore. If, on pouring off this, slag, a,
fresh quantity of ore be charged, the reaction will be very intense
because of the excess of oxygen present, and the time will be
proportionally hastened.
The process is carried out in two furnaces. In the first the
quantity of ore is deficient, and the metal after the reaction is
run into the second or finishing furnace, where the ore previously
heated is in excess. It must be so adjusted that the length of
time taken in the two operations should correspond, in order to
insure continuous working. The process is based upon rational
principles, inasmuch as the conditions are such as to favor de-
phosphorization at the lower temperature of the primary fur-
nace, and decarburization takes place at the higher temperature
of the secondary furnace. The slag from the primary furnace^
being richer in phosphoric acid, is of greater value for agricul-
tural purposes. The details of the process have often been
described. The author considers that when conditions admit
of obtaining cheaply a pig iron containing little silicon, from
1.7 to 2 per cent of phosphorus, 1.5 to 2 per cent of manganese,
and where a large production is desirable, the basic-Bessemer
process is to be preferred, as being less costly.
N^civ Opcii-?I earth Siccl Process 331
Dual processes have their uses when local circumstances
require the treatment of pig iron too high in silicon for basic
processes and too high in phosphorus for acid Bessemer practice.
xV typical system, which must, however, be regarded as a make-
shift, is that in which the pig iron is partially refined in an acid
converter and finished in a basic open-hearth furnace. It has
been employed at Sosnovice.
The Talbot Process. — This consists of working down a
charge in the ordinary way, withdrawing 25 to 35 per cent of
the bath into a ladle where it is recarburized and tapped, and
filling the furnace up with liquid pig iron. The bath then con-
sists of fluid metal, the carbon percentage of which has been
already lowered, and, on the addition of molten pig iron, the
strong reaction occurs. This reaction is stronger in proportion
as the difference between the amount of carbon in the added
metal and in the bath is greater, and hence hastens the opera-
tion, and renders it much quicker than the ordinary Siemens-
Martin process.
The furnace employed is a tilting one of the Wellman type,
and is too well-known to need detailed description. It is some-
what costly, amounting to £20,000 for a loo-ton furnace, but
as this estimate includes the cost of charging appliances and
accessories which can serve several furnaces, it may be more
correctly estimated at about £14,000 per furnace. The process
is in operation in America, at Pencoyd, where it originated,
and where the first furnace of 160 tons capacity made a weekly
•output of about 1,300 tons. It has also been introduced at
Frodingham and at Cardiff.*
The Talbot process yields a more regular and larger supply
of steel than the ordinary method, although its first cost is
greater. It is economical from a fuel convSumption point of
view, and the steel made is cheaper than ordinary Siemens steel,
although n6t so cheap as Bessemer steel. A disadvantage of
the process is the difficulty of making hard steel by it. The
steel can, it is true, be recarburized to any desired extent in the
ladle, but the method is uncertain owing to the losses of coke,
which cannot with certainty be controlled.
The modified Talbot process is in operation at Czenstochowa,
* It will also be adopted at the reconstructed Cargo Fleet Iron-
^\'orks. (Note by translator.)
332
The Iron and Steel Magazine
and was described in a paper by M. Surzycki, read at the recent
meeting of the Iron and Steel Institute. It seems to possess
the merit of being economical, but at the same time shares the
disadvantage already pointed out with regard to the homo-
geneity and quality of hard steel. It is difficult to regulate
the amount of metal tapped, and hence to determine with
accurac}^ the necessary amounts of material to be added to
recarburize.
Conclusions
The following broad conclusions may be drawn ;
Charging open-hearth furnaces with liquid pig iron is so
convenient and economical that there is no doubt it should be
adopted, and that with this view new open -hearth plants should
always be erected in the vicinity of blast furnaces. The ad-
vantages arising from this selection of site are:
1. Economy of labor in handling the output of the blast,
furnaces and in charging the steel furnace.
2. The time of working a charge being diminished by using
liqtiid pig iron, production per furnace can be increased.
3. Manufacture of steel is independent of the quality or
cost of scrap, of which the percentage proportion per charge can
be adjusted to suit the supply.
4. The steel works can reap the advantage accruing from,
the use of the waste blast-furnace gases for power purposes.
5. The fuel economy is considerable.
When ores containing not too much phosphorus are available,,
as at Jurjewka, the ordinary process of charging liquid pig iron
may be followed; with ores higher in phosphorus, but contain-
ing less than 1.8 per cent, the Bertrand-Thiel process may be
tried; when a mild steel is required, the Talbot and Surzycki.
processes give good results. As long, however, as ores — and,
consequently, pig iron — high in phosphorus are found abun-
dantlv, open-hearth processes in general will have a worthx'-
rival in the basic-Bessemer process.
TJic Cleaning of B last-Furnace Gas
333
THE CLEANING OF BLAST-FURNACE GAS*
By AXEL SAHLIN, London
'TpHE rapid development of the gas-motor during the last
'*' five years has given new value and importance to the gas
escaping from the blast fur-
nace, previously often de-
scribed as waste gas. This
" waste gas " has now become
a potential source of energy,
which, rightly used and hus-
banded, should, together with
the gas from the coke-ovens
supplying the blast furnaces,
suffice for the carrying out of
the entire series of convert-
ing and finishing processes
which transform the ore into
marketable steel products.
The gas leaving the blast
furnace carries with it a varying amount of gritty dust, which
has proved a more serious obstacle to the successful operation
of large gas engines than any mechanical imperfection in the
construction of these engines.
Successful efforts to remove the dust from the gas used in
the gas engine have, in a practical manner, demonstrated how
wasteful and imperfect have been our previous methods of
utilizing the valuable blast-furnace gas.
Until the appearance of the gas engine at the blast-furnace
plant the dust problem was, as a general rule, dealt with in a
different manner by each of the three large iron-producing
countries.
In England, in at least 90 per cent of the blast-furnace Avorks,
the dust in the gas was disregarded. A single downcomer
carried the gas escaping from the tunnel-head into an under-
ground gas-flue leading to stoves and boilers. In this flue and
in the stoves and boilers the bulk of the dust was deposited.
Every few months a rather long stoppage, regularly attended
* Read at the May, 1905, meeting of the Iron and Steel Institute.
J34 ^^^^ Iron and Steel Magazine
by vsubsequent more or less costly disturbances of operation,
was required for the cleaning of these fiues, while the cleaning
of stoves and boilers belonged to the daily routine of the works.
English engineers were in a position to ignore the dust problem,
partly because of the customary slow driving, and partly because
of the clean and firm structure of the materials used. But
even w^ith these advantages, stoves, and especially boilers, had
often to be put out of operation, cooled down and cleaned, and
the efficiency of the plant suffered.
In Germany the majority of the furnaces were equipped
with a dry or wet dust-catcher, consisting of a series of upright
tabes of small diameter, the open bottoms of which extended
downwards into a common water-seal. The dust was deposited
mostly by centrifugal force, as the direction of the flow of the
gas was suddenly changed in passing through the cross tubes
connecting one chamber with the other. Usually six of these
chambers from 6 feet to 8 feet in diameter, and often 60 feet in
height, constituted the dust-catcher. The collected dust was
raked from the pan forming the water-seal by manual labor.
The gas leaving the dust-catcher could not be considered prop-
erly cleaned.
Since the discovery of the Mesabi mines American blast-
furnace engineers have been compelled to use a burden contain-
ing from 50 to 100 per cent of dust-fine ore, and the question
of cleaning the gas became one of imperative importance.
The generally adopted American dust-catcher consists of a
wide, unobstructed chamber, through which the gas passes
and in which its velocity is greatly retarded. The bottom of
the dust-catcher is built as a hopper, closed at its lowest point
by a door or valve, through which the dust from time to time
is removed Vjy gravity.
It required the appearance of the gas engine to compel
improvements in these methods, and to stop, at least partly,
the waste which we all complacently had permitted to go on
from year to year.
We have now learned and realize that the whole of the gas
escaping from the blast furnace should be utilized for subse-
quent refining and finishing processes, and that, before being
so used, it should be thoroughly cleaned.
Bv cleaning all the gas we would save:
The (leaning of Blast-Furnacc Gas 335
1. One stove per furnace. No reserve will thenceforth be
necessary to maintain the capacity of the furnace during the
periods of cleaning of the different stoves, and three stoves
per furnace will give good practice.
2. The heating surface of the three remaining stoves may
be made 10 per cent smaller than hitherto, because, with dusty
gas, the value of the heating surface will gradually decrease,
whereas with clean gas it will have a permanent efficiency.
3. The heating surface of the boiler plant may be reduced
from 10 to 15 per cent for the same reason, and because the
necessity for periodical cleaning of the boiler settings will
disappear.
4. The repairs to the boiler plant will be reduced as the
strains due to frequent cooling of the boilers are avoided.
5. The labor force previously constantly employed in
cleaning stoves and boilers may be dispensed with.
6. The more perfect combustion will reduce gas consump-
tion and liberate for use in other departments a considerable
percentage of gas hitherto wasted. The percentage of gas
thus saved may be estimated at about 20 per cent of the volume
burned in stoves and boilers.
7. The success of blast-furnace operations depends entirely
on the regularity and uniformity of conditions affecting the
furnace. Uniform fuel gas will give uniform heat and uniform
power, and will add increased efficiency to the furnace, which
we all recognize, though it may not be easy to express the gain
in exact figures.
Assuming, then, as an understood and accepted axiom, that
blast-furnace gas should be cleaned before being burned, we
proceed to consider the process of cleaning from two points
of view: (i) The degree of purity reached, and (2) the cost
of reaching it.
It is the removal of the last fraction of a gram of dust
contained in a cubic meter of gas which largely adds to the
expense of cleaning. For stoves, boilers, kilns, furnaces, etc.,
such a small quantity of exceedingly fine and ]:)Uoyant dust is
no detriment, as it will pass off with the products of combus-
tion through the smoke-stack. A gas containing 0.3 gram
per cubic millimeter, or less, is sufficiently clean for these pur-
poses.
336 The Iron and Steel Magazine
For the gas engine, on the other hand, the gas can never be
too pure, and the removal of the last tenth of one gram, though
expensive, will pay for itself in longer life and decreased repairs
to the expensive engines.
The cleaning of the blast-furnace gas should, therefore,
take place in three stages, as follows:
I I. The preliminary dry cleaning, which does not involve
extra operating expenditures.
2. The wet cleaning to fit the entire quality of gas for use
in stoves and under boilers, in roasting kilns, furnaces, etc.
3. The special cleaning for purifying part of the gas for
power purposes.
I . Dry Cleaning. — The gas should be removed from the
blast furnace as m.uch as possible symmetrically around the
circumference of the furnace, so as to avoid an excessive flow
of gas in any one direction. The use of four gas uptakes, which,
curving downwards, should unite into one main downcomer, is
to be recommended; this downcomer to enter the dust-catcher
tangentially and near the bottom of the chamber. It is a
mistake, from the point of view of cleaning, to make the down-
comer too wide. It should be designed so as to cause the gas
to flow with an average velocity of from 25 to 35 feet per second.
The dust-catcher should be built in the shape of a vertical
cylinder with conical top and bottom ; it should have a diameter
of not less than four times that of the downcomer. The de-
livery opening from the dust-catcher should be placed centrally
at the top of the same. The gas will, by this arrangement, on
entering the dust-catcher, be given a rapid swirling motion, and
the two powerful and inexpensive agents, gravity and centri-
fugal force, will combine to deposit in the dust-catcher all the
heavier particles of dust. According to the nature of the
materials charged, the dry dust thus collected may, or may not,
be worth briquetting and recharging.
The gas leaving a well-designed dust-catcher may still con-
tain from 2 to 8 grams of dust per cubic meter, depending largely
on the character of the ore employed and the rate of driving of
the furnace.
The whole volume of gas should next be submitted to the
second stage of the cleaning process.
2. Wet Cleaning. — The second cleaning should be effected
The Clcaiiiiig of Blast-Furnace Gas 337
by means of \vater. Experience has proved that the line dust
settles most readily on cool and moist surfaces. The passage
of gas through open chambers, even though profusely sprayed
with water, has not proved effective.
Inventive effort has, during the past five years, been directed
towards providing gas-cleaning apparatus, which may be
grouped into three classes:
(a) Stationary cleaners or scrubbers.
(6) Rapidly revolving or atomizing machines.
(c) Slowly revolving or fresh contact cleaners.
(a) Stationary Cleaners. — These consist of chambers from
the top of which sprays of water, distributed over the entire
section of the chamber, descend. The water is arrested by a
number of grids or porous masses of material with numerous
interstices. Iron bars, rolled and cast, tangled wire, coke
or wooden slats, have all been used with success as filling mate-
rial or grids. The gas enters the apj^aratus from below the
grids and ascends, meeting the sprays of water falling from
above through the innumerable interstices in the obstructing
grids or masses, to the wet surfaces of which the dust adheres,
only to be washed away by the falling drops of water. The
Zschocke cleaner is a well-designed type of such apparatus.
It is claimed that the gas leaves this apparatus with from
3.0 to 1.4 grams of dust per cubic meter.
(6) Rapidly Revolving Machines. — The earliest form of the
rapidly revolving or atomizing apparatus was, as a matter of
course, the fan. Having been in use for generations, it was
at once available as a mechanically perfect tool. The water
injected into the fan, together with the gas, was forcibly dashed
against the casing, removing the dust.
A modification of the fan is the Theisen apparatus, which
provides a forced contact between gas and spraying water
driven against a cylindrical envelope by fan blades set at oblique
angles to the axis of rotation.
There are probably no more effective contrivances known
to-day for elimination of the last fractions of a gram of dust
per cubic meter than a properly designed and rapidly revolv-
ing fan or a Theisen apparatus. But the good work is not
done cheaply, as the power consumed for revolving the blades
increases greatly when the water sprays are put in operation.
338 The Iron and Steel Magazine
This class of apparatus is, therefore, most suitable for thoroughly
cleaning the portion of gas used for generating power.
(c) Slowly Revolving Apparatus. — The considerable power
required to drive rapidly revolving water-sprayed fans, on the
one hand, and the difficulty of insuring a thorough and uni-
form cleaning of the grids or checkers of the stationary appa-
ratus on the other hand, led to the design of slowly revolving
gas cleaners, which are particularly suited for the second, or
wet, cleaning of the entire volume of gas produced by the
furnace before it is admitted to the stoves, boilers, kilns, or to
the fans which finally prepare it for the gas engines.
The Bian apparatus is the earliest form of this type, and
has proved very effective. It consists of a horizontal cylin-
drical casing, through which the gas is passed from end to
end. In the axis of the cylinder revolves slowly a shaft, to
which are bolted a number of circular disks of perforated plates
of a diamieter but little less than the inside diameter of the
casing. The gas is, therefore, in its progress compelled to
pass through the perforations in the disks. Between the per-
forated disks extend from the outer envelope annular deflect-
ing plates, which compel the gas to approach the center of
the apparatus after each passage through the perforations in the
revolving disks. The outer shell is filled with water up to the
bottom of the central shaft. In revolving the perforations in
the disks are therefore dipped into water and thoroughly washed
for every revolution of the apparatus, and ascend from the bath
covered with thin films of water. The dust which settles on
the bottom of the apparatus is from time to time drained off
through suitable sludge valves.
This apparatus is effective, and requires but little power
slowly to revolve the central shaft and the disks fastened thereon.
With the view of securing larger capacity of each apparatus
(that is, a larger area of perforations in proportion to the diam-
eter of the apparatus) , a better and more continuous wetting
of the perforated surfaces and a more automatic and constant
removal of the dust, my firm have designed a new style of
slowly rotating gas-cleaning apparatus, which has been named
the Sahlin revolving gas-cleaner, shown by Fig. i.
It consists, like the Bian apparatus, of a horizontal cylin-
drical shell. Parallel with the axis of this shell, but somewhat
Fig.
340 The Iron and Steel Magazine
below the center of the same, is placed a horizontal shaft. On
this shaft are fitted, alternately, spider arms supporting rings
of double 3-inch angle iron riveted together, and solid plate disks
carrying at their circumference similar angle rings. Between
the angles of the spider arms and those of the disks are bolted
perforated plates forming the envelope of a drum. The oblong
perforations are wider at the end of the cylinder where the gas
enters, and gradually narrower towards the discharge end.
Between the outer shell and the inner revolving drum is formed
a horseshoe-shaped space, widest at the top. Closely behind the
rim of each spider, this space is divided by plate diaphragms
riveted to the outer shell, and practically closing the open area
between shell and drum. To the angle rings of the frame, sup-
ported by the spider arms and by the disks, are bolted a series of
spirallv bent flat bars, or scrapers, which closely approach the
bottom of the shell. Near the gas inlet is riveted to the lower
part of the shell a water-sealed discharge opening. At the
bottom of the shell, near the gas entrance, a drain-cock or sludge
valve is provided for emptying the apparatus. Along the top
of the shell three rows of sprinklers are arranged, so as to deluge
the entire top of the revolving perforated plates with a dense
spray of water. The drum^ is turned by worm and worm gear
through a belt pulley driven from the ventilator described below.
When gas enters the apparatus through the head of the outer
shell, it is admiitted into the drum between the spider arms sup-
porting the first angle iron ring, but is then confronted by the
first solid disk. The only passage offered is through the narrow
and dripping perforations of the drum. In passing these, the
gas is met by the profuse spray from the sprinklers. It is at
the same time spread out into a thin sheet in the horseshoe-
shaped space between the two cylinders, and advances outside
and beyond the first revolving disk until met by the first horse-
shoe-shaped diaphragm, which prevents further progress and
compels the gas again to enter the interior of the drum beyond
the first disk. Passing through the openings in the second spider
wheel, it is again compelled to flow outward through the per-
forations in front of the second disk, and so on, the gas passing
forward and back through the perforated envelope, now spread-
ing itself into a thin layer, now contracting itself into a cylin-
drical column, and continuously exposed to the intense spray
The Clcaiiiiif: of B last-Furnace Gas 341
from the sprinklers. The gas has, of course, to turn an angle of
180° for every time it passes through the perforations. It is
finallv delivered through the discharge opening, which corre-
sponds to the opening of admission, at the opposite end of the
shell.
The dust is settled in the water at the bottom of the shell
and is moved by the spirally-bent scrapers towards the inlet
end, where it is discharged automatically through the water-
sealed overflow.
As the areas of drum, perforations and horseshoe -shaped
space are all very large, compared with the section of the down-
comer, the velocity of the gas in passing through the apparatus-
is greatly reduced. The resistance offered by the apparatus
is, therefore, insignificant, even though the entire volume of
gas produced by one furnace is passed through a single apparatus.
The perforated drum is revolved about six reversions per
minute, requiring less than ij horse-power, which power is
transmitted by a belt from the shaft of the ventilator.
It has been found that whether a stationary or slowly revolv-
ing apparatus is used for the second cleaning, the tunnel-head
pressure must be assisted by a ventilator, to insure an even flow
of gas. This ventilator is, however, by no means as great a
consumer of power as would be water-sprayed cleaning fans or
atomizers. A ventilator passing 40,000 cubic feet of gas per
minute (equal to about 70,000 cubic meters per hour) against
a head of 2 inches water column, requires only about 35 horse-
power.
Safety Device. — If for any reason, such as the hanging of
the furnace or the sudden stoppage of the blowing engine, the
fxOw of gas from the furnace should cease or be reduced, there
is with all forms of gas-cleaning apparatus a danger that air
may be drawn into the apparatus and gas tubes, where the
mixture of gases formed may, in unfavorable circumstances,
cause serious explosions. To prevent the possibility of this, I
recommend a very simple and effective device, consisting of
a small gasometer having a diameter of 8 feet, with a lift of
6 inches. This gasometer is so balanced that it will drop when
the pressure inside the piping falls to J-inch water column.
The interior of the gasometer is connected by a pipe with
the dry dust-catcher. An arm projecting from the top of
342
The Iron and Steel Magazine
gasometer connects with a switch in the circuit carrying the
electric current to the motor driving the ventilator and cleaner
in such manner that, when the gasometer drops, the current is
broken and the motor stopped. When the returning pressure
in the dust-catcher causes the gasometer to rise, the circuit is
automatically closed and the cleaning apparatus started.
Fig. 2
A general outline of a cleaning plant for the entire volume
of gas from one blast furnace is shown by Fig. 2.
The gas leaves the furnace through four uptakes placed at
90" angle. These join into one downcomer, which delivers
the gas in tangential direction into a cylindrical dust-catcher.
From the top of the dust-catcher the gas passes through one
The Clean iin^ of Biast-Funiacc Gas 345
of two mushroom valves. The valve nearest to the stoves con-
nects the dust-catcher direct with the gas main, and is opened
only in case of repairs to the gas-cleaning plant. The second
valve connects the dust-catcher with a water seal, from which
the gas flows into the Sahlin gas cleaner described above. From
this gas cleaner it reaches a ventilator which delivers the now
sufficiently clean and cooled gas into the bottom of a dryer. By
closing the drain pipe and admitting water, this scrubber or
dryer miav be changed into a second water-seal, effectively
isolating the cleaner. Crossbars inside the dryer carry a column
of coarse coke. In passing through this the gas is freed from
the bulk of the water which may be mechanically carried from
the second or wet cleaning. At the top of the dryer a mush-
room valve again admits the gas into the general gas main.
There is no spare or reserve provided for this simple wet gas-
cleaning plant. The damage caused by occasionally, for short
periods, using gas direct from the dust-catcher, as now is being
done continuously, is not sufficient to justify an increased invest-
ment in duplicating the plant.
The power required for the second stage of cleaning of
40,000 cubic feet of gas of a temperature of 20° C. is, as above
stated, 36^ horse-power, to which must be added 8 horse-
power for pumping of cooling water.
3. Gas for the Power Plant is drawn from the clean gas.
main and is passed through one of two electrically driven fans
sprayed with water and discharging into a second smaller dryer,,
whence the gas, now% practically speaking, dust free, is sent
to the engines. The size of this plant depends on the quantity
of gas required for power purposes. A fan cleaning 10,000
cubic feet per minute, and using 8,000 gallons of spraying water
per hour, requires about 65 horse-power, and will supply gas
engines of from 5,000 to 6,000 horse-power. The pumping of
spraying water will require about 3 horse-power additional.
The cooling water used in the cleaning processes is delivered
into double settling and cooling ponds, and is thence lifted by
centrifugal pumps into a standpipe or water-tank, from which
it is returned by gravity to the cooling plant. On top of the
tank may be arranged trays for additional cooling of the water„
Slightly modified, as shown b}^ Fig. 3, the secondary clean-
ing plant may be attached to any existing gas main. A slide
344
The Iron and Steel Magazine
valve is inserted into the gas main, and on either side of this
valve is placed, on top of the main, a mushroom valve. The
mushroom valve on the side nearest the furnace will connect
•directly with the water-seal and the Sahlin gas cleaner. The
mushroom valve on the side away from the furnace will return
the gas directly from the dryer into the existing gas main.
Should it be necessary to stop the gas-cleaning plant, the slide
0 tXISTINC CAS MAIN I
EAISTinC CAS
IS MAIN )
Fig.
valve is lifted and the mushroom valves are closed, the gas,
according to present general practice, passing direct!}^ from the
dry dust-catcher to stoves and boilers, or to the special clean-
ing plant for the gas engines.
Interruptions in the working of the secondary cleaning plant
will, however, be infrequent, as the simplicity and solid con-
struction of the plant will reduce repairs to a minimum.
M
ABSTRACTS *
{From recent articles of interest to the Iron and Steel Metallurgist)
ICRO-METALLOGRAPHY with Practical Demonstration.
J. E. Stead. " Journal of the Royal Microscopical So-
ciety. 7,500 w., illustrated. — In the first part of this paper
the author describes the technology of metallography while the
second part is devoted to a description of methods for detecting
the more highly phosphorized portion of iron and steel. The
following are abstracts from this paper:
Methods jar Detecting the more Highly Phosphorized Por-
tions in Iron and Steel. — On reading the published researches-
of micro-metallographers it would appear that very little atten-
tion has been paid to the methods for detecting or identifying
the more highly phosphorized portions in iron and steel. I have,
however, repeatedly had occasion to report upon the structure
of steels and to draw attention to irregular distribution of
phosphorus.
I have already published the methods of detecting phos-
phide in pig irons by the microscope ; and it only remains for me
to describe other methods for differentiating between the por-
tions higher and lower in phosphorus in commercial irons and
steels.
The following are detailed directions for applying the several
methods :
Heat Tinting Methods. — When polished iron or steel is-
heated in air the surface becomes colored by the formation of
* Note. The publishers will endeavor to supply upon request the full
text of the articles here abstracted, together wilh all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60-
cents, "D" 80 cents, "E" $1.00, "F" $1.20, "G" $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cases
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
ing upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.
345
346 The Iron and Steel Magazine
films of oxide of iron. In proportion as the temperature is
raised, or continued at one suitable temperature, the tints pass
from pale yellow to yellow, brown, purple, blue and steel gray,
and through the same series of tints a second time if the heating
is continued, but the tints of the second series are not so intense
as those of the first.
Massive carbide of iron becomes colored less rapidly than
iron and more rapidly than phosphide of iron, whilst iron con-
taining phosphorus in solid solution colors more rapidly than
pure iron or iron containing less phosphorus.
Method I. — Into an iron crucible or ladle, or other suit-
able receptacle, is placed about 4 ounces of tinman's solder(2 tin,
I lead). The vessel is placed over a Bunsen burner and the
solder melted. Into the metal a Le Chatelier couple, covered
with a thin piece of asbestos paper, is inserted. The flame of
the burner is adjusted until the temperature of the metal stands
at 250° C. The specimens, having been polished, are rubbed with
a piece of clean woolen cloth, and are warmed on a hot plate, or
in a boiling-water oven, and when still warm the}^ are again
rubbed with the cloth. They are then floated on the molten
metal. The reason for first gently heating is to prevent conden-
sation of acid water from the waste products of the burning gas.
If the precaution is not taken the specimen after heating will be
covered with minute colored dots due to condensed steam. The
surfaces of the specimens are watched and examined with a
strong magnifying glass. They will assume a regular yellow
tint, and in a few minutes the phosphorized portions will become
brown on a yellow ground, and if the heating is continued they
will become colored blue, whilst the parts not so high in phos-
phorus will be brown or dark yellow. At this point the speci-
miens are removed and may be examined under the microscope
whilst still hot. If the tinting is not sufficiently advanced they
may be returned to the bath for further heating.
Method 2 . — Instead of regulating the temperature of the
bath it may be heated until the surface of the solder begins to
form yellow films. Each specimen, preferably of the dimen-
sions 20 mm. by 10 mm. by 5 mm., is, after warming and rub-
bing with a cloth, held at one end with a pair of tongs, and the
under surface of the other end is immersed in the highly heated
metal. In one minute or less the tinting will be complete, but
Abstracts 347
it will be graduated in color between gray at one end and pale
vellow at the other; the intermediate part passing through the
whole gamut of coloring. The specimens are removed when the
central parts have assumed a brown color.
Treated in this way the phosphorized portions will be dark
brown on a yellow ground, or blue on a brown ground.
Method J. — The specimen is heated rapidly until uniformly
blue, and when cold is immersed in water containing a one-
thousandth part of nitric acid. The films covering the phos-
phorized parts will be dissolved in advance, and if the acid treat-
ment is stopped at the right moment it is possible to have white
phosphorized areas on a brown or blue matrix. This method
gives very satisfactory results, but many failures to obtain the
exact development may follow the first attempts. It is some-
times advisable to rub the developed specimen with moistened
chamois leather before drying with a hot blast of air.
Method 4. — Instead of floating the specimens on the sur-
face of liquid metal, they are placed in a jacketed copper
chamber 4 inches in length and i inch square, which is sur-
rounded, excepting at one end, with heavy mineral oil, main-
tained at a temperature of 245° C. A drawer is fitted into this,
and into it the metal sections are placed. The tinting by this
method of heating is more under control than by the first de-
scribed, and it is easy to locate the parts highest in phosphorus
even in steel castings containing under .05 per cent of that
element.
Iodine Etching. — This method is based on the fact that a
very dilute tincture of iodine in potassium iodide corrodes the
portions lower in phosphorus relatively more rapidly than those
containing more of that element.
The necessary reagent contains i gram of iodide and .01
iodine per 500 c.cm. alcohol and 50 c.cm. water.
The polished specimens are immersed in this and are ex-
amined from time to time. When it is seen that some portions
remain brilliantly white on a dull ground, they are removed,
washed with water and alcohol, and dried in a current of
hot air.
In longitudinal sections of rolled steel after this treatment
there will be seen white lines which may or may not be indepen-
dent of the ferrite and pearlite areas. These white lines con-
348 The Iron and Steel Magazine
tain the higher proportion of phosphorus. Relatively they
resist the corrosive action of the iodine. That this is so may be
verified by a longer action followed by slight re-polishing on wet
parchment, when, even with the aid of a simple lens, the resist
lines will be seen to stand in relief. When examined under
oblique light rays, the phosphorized parts appear black on a
light ground.
Picric Acid Etching and Tinting Method. — The long-
continued action of a 2 per cent solution of picric acid in water
containing 5 per cent alcohol will color the portions higher in
phosphorus, yellow, brown, blue, etc.
This method is well adapted for the study of wrought iron
and soft steel.
When applying the reagent the specimens are immersed in
the solution.
The coloring may take several minutes to develop. When
it is considered advisable to remove the specimens, they must be
washed with water and alcohol, dried in a current of hot air, and
on no account must they be wiped with a cloth, for the slightest
friction is liable to remove some of the films.
A simple solution of picric acid in water colors the phos-
phorized portions in advance of the parts containing less phos-
phorus, but all parts will eventually become brown if the action
is continued long enough.
Nitric Acid Etching and Tinting Method. — This method
is based on the observation that very dilute nitric acid, like
iodine, acts relatively less rapidly on the phosphorized portions,
and at first they remain bright, but, if the action is continued,
they become darkened by the formation of a dark-colored skin
or film. This film is probably of the same substance as the
black residue which remains when phosphorized steels are dis-
solved in dilute sulphuric acid.
On etching longitudinal sections of steel and iron, the phos-
phorized lines at first resist the acid and appear white on a dark
ground, but after longer action the white lines become relatively
darker than the less phosphorized parts — indeed, it is possible
with care to obtain a positive and negative appearance on the
same specimen by a short or more prolonged etching.
After strong etching, if the specimen is repolished on a
•cloth block, the phosphorus lines will stand in relief, and as the
Abstracts 349
dark stain is readily removed by slight friction, the lines appear
white on a dull ground.
Professor Heyn has kindly sent me some photographs'of steel
structures developed by his copper-ammonium-chloride reagent,
which appeared to be identical with those developed by iodine.
Although he does not describe them as other than indicative of
primary crystallization, I have but little doubt that they are
mainly evidence of imperfect distribution of phosphorus.
If steel, containing low carbon, say under .5 per cent, in
either the cast or forged condition, is very slowly cooled, the
highh^ phosphorized areas reject the carbon which has segre-
gated with the phosphorus, and as a result massive areas of
ferrite appear, the borders of which are often surrounded with
pearlite.
If the phosphorus is greatly concentrated in certain parts,
carbon will not be retained there even on comparatively rapid
cooling from a high temperature. No. 413.
Corrosion of Boiler Tubes. Rear- Admiral John D. Ford,
U. S. N. " Journal of the American Society of Naval Engi-
neers." 10,000 w., illustrated. — In this report the author de-
scribes the results of his investigation to ascertain the rate of
corrosion of steel and iron boiler tubes. According to direction
from the Bureau of Steam Engineering of the Navy Depart-
ment the tests were made in twelve tanks, each tank containing
sixteen samples from top, middle and bottom of steel ingots.
The samples were shifted in their position each week, so that
at the end of a period of sixteen weeks each sample had occupied
consecutively every position in the tank in which it was placed,
and at the end of each such period the samples were taken from
the tanks, cleaned with a brush (but not scraped), washed, dried,
weighed and photographed. The present report is based upon
the results obtained at the expiration of the first, second, third
and fourth periods of sixteen weeks. The tests were conducted
in a room adjoining the chemical laboratory of the National
Tube Company, that company having offered to furnish the tubes
and give freely all necessary facilities and assistance. Both
cold- and hot-drawn tubes were tested. The average loss in
grams after each period of sixteen weeks is given below:
35© The Iron and Steel Magazine
Average Loss in Grams
After 1 6 After 32 After 48 After 64.
weeks weeks weeks weeks
Class of ttibe (ist 16 (2d 16 (3d t6 (4th 16
weeks) weeks) weeks) weeks)
(i) Hot-drawn seamless steel. . . .3034 -5333 -351° -S^T^
(2) Lap-welded Bessemer steel . .3147 -495° -4043 •494S
(3) Cold-drawn seamless steel . . .3268 -5795 -4595 -S^J^
(4) Charcoal iron -3326 -5896 -3966 -4893
The difference in the amount of loss in the four classes of
tubes is not great, that of the charcoal-iron tubes, which had the
greatest loss, being, roughly, 6.3 per cent more than the loss of
the hot-drawn seamless tubes, which suffered the least.
Considering the three grades of the hot-drawn seamless
tubes alone, it was found that the average losses per square
inch of surface area were:
After 16
weeks
(ist 1 6
weeks)
After 32
weeks
(2d 16
weeks)
After 48
weeks
(3d 16
weeks)
After 64
weeks
(4th 16
weeks)
•3205
.5700
.3708
•5213
.3121
•5339
•3571
.5108
.2776
.4960
•3254
.4906
Tubes from the top of ingot. . . .
Tubes from the middle of ingot .
Tubes from the bottom of ingot
The difference in the losses of these three grades is very
marked. At the end of the first sixteen weeks those from the
top of the ingot lost 15.4 per cent more, and those from the
middle of the ingot 12.4 per cent more, than those from the bot-
tom of the ingot. At the end of thirty-two weeks those from
the top of the ingot lost 14.9 per cent more, and those from the
middle of the ingot 7.6 per cent more, than those from the bottom
of the ingot. At the end of forty-eight weeks those from the
top of the ingot lost 15.9 per cent more, and those from the
middle of the ingot 9.7 per cent more, than those from the bot-
tom of the ingot. At the end of sixty-four weeks those from the
top of the ingot lost 6.2 per cent more, and those from the middle
of the ingot 4.1 per cent more, than those from the bottom of the
ingot; the loss from the bottom samples being less, probably,,
on account of the lesser amount of impurities and greater density
of the metal of the ingot in this part.
Of the three grades of lap-welded Bessemer tubes the aver-
age losses per square inch were:
Abstracts
351
After lO
weeks
(ist 1 6
weeks)
After 32
weeks
(2d 16
weeks)
After 48
weeks
(3d 16
weeks)
After 64
weeks
(4th 16
weeks)
.. .3686
.5268
.4112
.4966
it. .2941
.4826
•3939
.4891
^ot .2814
•4756
.4079
•4977
Tubes from the top of ingot.
Tubes from the middle of ingot
Tubes from the bottom of ingot
The difference here is even more marked for the top of the
ingot than in the case of the hot-drawn tubes. At the end of
the first sixteen weeks those from the top of the ingot lost 30.98
per cent more, and those from the middle of the ingot 4.51 per
cent more, than those from the bottom of the ingot. At the end
of thirty-two weeks those from the top of the ingot lost 10.76
per cent more, and those from the middle of the ingot 1.47 per
cent more, than those from the bottom of the ingot. At the
end of forty-eight weeks those from the top of the ingot lost
.8 per cent more, and those from the middle of the ingot 3.43
per cent less, than those from the bottom of the ingot. At the
end of sixty-four weeks those from the top of the ingot lost .22
per cent less, and those from the middle of the ingot 1.72 per cent
less, than those from the bottom of the ingot; the great varia-
tion in loss of these samples being due probably to a greater
amount of impurities of the metal of the ingot in this heat.
When the results obtained from the cold-drawn seamless
tubes are investigated, a similar result as in the former cases
appears, except that the losses in each series of sixteen weeks is
much smaller than in the other cases.
The losses per square inch of area were :
After 16
weeks
(ist 16
weeks)
After 3 2
weeks
(2d 16
weeks)
After 48
weeks
(3d 16
weeks)
After 64
weeks
(4th 16
weeks)
.3288
.6166
.4796
•5525
.3120
•5278
•4451
.4498
•3075
•5645
•4451
•5192
Tubes from the top of ingot. . . .
Tubes from the middle of ingot .
Tubes from the bottom of ingot
The difference in the losses of these three grades is quite
marked; at the end of the first sixteen weeks those from the
top of the ingot lost 6.9 per cent more, and those from the mid-
dle of the ingot 1.4 per cent more, than those from the bottom of
the ingot; at the end of thirty-two weeks those from the top
of the ingot lost 9.2 per cent more, and those from the middle of
the ingot lost 6.49 per cent less, than those from the bottom
of the ingot; at the end of forty-eight weeks those from the
35 2 The Iron and Steel Magazine
top of the ingot lost 7.7 per cent more than, and those from
the middle of the ingot were equal to, those from the bottom of
the ingot ; at the end of sixty-four weeks those from the top of the
ingot lost 6.4 per cent more, while those from the middle of the
ingot lost 13.3 per cent less, than those from the bottom of
the ingot. No. 414.
Sulphur in Coke and its Behavior in the Blast Furnace.
F. Wuest and F. Wolff. Iron and Steel Institute, May,
1905. 13,000 w. — The authors describe some experiments,
conducted to ascertain the behavior of sulphur in the blast
furnace consisting chiefly in the determination of the combusti-
ble sulphur and in the reaction of that element when brought in
contact at a high temperature with hydrogen, steam, nitrogen,,
carbon monoxide and carbon dioxide. From their results they
draw the following conclusions:
" Contrary to the generally held opinion, the sulphur in the
coke does not reach the level of the tuyeres of the blast furnace
without undergoing alteration, but a great portion of it is pre-
viously volatilized by the ascending gases, and is then largely
absorbed from the gases by the descending charge, and in this
condition arrives in front of the tuyeres.
" Up to 800 degrees the sulphur is principally absorbed by
the oxides of iron from the sulphur-laden gases, while from 8oo-
degrees upwards the position is reversed, and the lime becomes
the chief absorbent of the sulphur." No. 415.
Britain's Earliest Iron Furnaces and Molding Floors.
Thomas May. " The Iron and Coal Trades Review," August 11,.
1905. 10,000 w., illustrated. — The author describes with
numerous illustrations the Roman methods of manufacturing"
iron as evidenced by the remains of smelting, molding and other
furnaces discovered during recent excavations which he con-
ducted within the Roman fortifications at Wilderspool, near
Warrington. No. 416. A.
The Testing of Cast Iron. Richard Moldenke. " Journal
of the Franklin Institute," June, 1905. 5,000 w. — The author
reviews the progress made in the testing of cast iron, with special
reference to American practice. No. 417. C.
Abstracts 353
The Bertrand-Thiel Process. E. von Maltitz. " The Iron
Age," August 10, 1905. 9,000 w. — The author describes the
Bertrand-Thiel process as conducted at Kladno and elsewhere
and shows how it can be introduced with good results in almost
any open-hearth steel plant. No. 418. A.
METALLURGICAL NOTES AND COMMENTS
c,. T X1-- -o 11 * Sir Lowthian Bell, who died on December 20,
Sir Lowthian Bell *
1904, was bom in Newcastle on February 15,
1816, and educated, first at Bruce's Academy in Newcastle, and
afterwards in Germany, in Denmark, at Edinburgh University
and at the Sorbonne, Paris.
On the completion of his studies, Lowthian Bell joined his
father at the Walker Iron Works. Mr. John Vaughan, who
was with the firm, left about the year 1840, and in conjunction
with Mr. Bolckow began their great iron manufacturing enter-
prise at Middlesbrough. Mr. Bell then became manager at
Walker, and blast furnaces were erected under his direction.
He became greatly interested in the ironstone district of Cleve-
land, and as early as 1843 niade experiments with the ironstone.
He met with discouragements at first, but was rewarded with
success later, and to Messrs. Bell Brothers largely belongs the
credit of developing the ironstone field of Cleveland. Mr. Bell's
father died in 1845, ^^^ "the son became managing partner. In
1852, two years after the discovery of the Cleveland ironstone,
the firm acquired ironstone royalties first at Normanby and
then at Skelton in Cleveland, and started the Clarence Iron
Works, opposite Middlesbrough. The three blast furnaces here
erected in 1853 were at that time the largest in the kingdom,
each being 47^ feet high, with a capacity of 6,012 cubic feet.
Later furnaces were successively increased up to a height of
80 feet in 1873, with 17 feet to 25 feet in diameter at the bosh,
8 feet at the hearth, and about 25,500 cubic feet capacity.
Sir Lowthian earned great repute as an author. He was
a prolific writer on both technical and commercial questions
relating to the iron and steel industries. His first important
book was published in 1872, and was entitled " Chemical Phe-
nomena of Iron vSmelting: An Experimental and Practical Ex-
* Extracted from "The Journal of the Iron and Steel Institute,"
No. XI, 1904.
354
I
Metallurgical Notes and Comments 355
amination of the Circumstances which Determine the Capacity
uf the Blast Furnace, the Temperature of the Air, and the Proper
Condition of the Materials to be Operated Upon." This book,
which contained nearly 500 pages, with many diagrams, was the
direct outcome of a controversy with the late Mr. Charles Coch-
rane, and gave details of nearly 900 experiments carried out
over a series of years with a view to finding out the laws which
regulate the process of iron smelting, and the nature of the
reactions which take place among the substances dealt with in
the manufacture of pig iron. The behavior of furnaces under
varying conditions was detailed. The book was a monument
of patient research which all practical men could appreciate.
His other large work, covering 750 pages, was entitled, " The
Principles of the Manufacture of Iron and Steel." It was issued
in 1884, and in it the avithor compared the resources exivSting
in different localities in Europe and America as iron-making
centers. His further investigations into the manufacture of
pig iron were detailed, as well as those relating to the manufac-
ture of finished iron and steel.
In 1886, at the instance of the British Iron Trade Association,
of which he was then president, he prepared and published a
book entitled, " The Iron Trade of the United Kingdom Com-
pared with Other Chief Iron-Making Nations." Besides these
books and numerous papers contributed to scientific societies,
Sir Lowthian wrote more than one pamphlet relating to the
history and development of the industries of Cleveland.
In 1876, Sir Lowthian was appointed a Royal Commissioner
to the Centennial Exhibition at Philadelphia, and wrote the
official report relating to the iron and steel industries. This
was issued in the form of a bulky Blue-Book.
As a director of the North-Eastern Railway Company, Sir
Lowthian prepared an important volume of statistics for the use
of his colleagues, and conducted exhaustive investigations into
the life of a steel rail.
The majority of his papers were read before the Iron and
Steel Institute.
To him came in due course honors of all kinds. When the
Bessemer Gold Medal was instituted in 1874, Sir Lowthian was
the first recipient. In 1895, ^^ received at the hands of the
King, then Prince of Wales, the Albert Medal of the Society of
356 The Iron and Steel Magazine
Arts, in recognition of the services he had rendered to arts,
manufactures and commerce by his metallurgical researches.
From the French government he received the cross of the Legion
of Honor. From the Institution of Civil Engineers he received
the George Stephenson Medal in 1900, and, in 189 1, the Howard
Quinquennial Prize which is awarded periodically to the author
of a treatise on iron.
For his scientific work Sir Lowthian was honored by many
of the learned societies of Europe and America. He was elected
a Fellow of the Royal Society in 1875. He was an Honorary
D.C.L. of Durham University; an LL.D. of the Universities of
Edinburgh and Dublin; and a D.Sc. of Leeds University. He
was one of the most active promoters of the Durham College of
Science by speech as well as by purse; his last contribution was
made only a short time ago, and was ;^3,ooo, for the purpose of
building a tower. He had held the presidency of the North of
England Institution of Mining and Mechanical Engineers, and
was the first president of the Newcastle Chemical Society.
In the Iron and Steel Institute he took special interest.
One of its original founders, in 1869, he filled the office of presi-
dent from 1873 "to 1875.
Cast-iron Wheels. — The discussion which has been going
on recently over the use of cast-iron wheels under high-capacity
cars differs somewhat from previous discussions, which have
taken place periodically during the last twenty years, in that the
purely mechanical problem of producing a wheel of chilled cast
iron which is safe and which will give good service has been lost
sight of. The discussion has hinged on the commercial problem
of whether or not a cast-iron wheel can be made for a certain
price and at the same time be safe and give long service. The
mechanical problems involved in the' design and manufacture
of cast-iron wheels for freight cars are more difficult than they
have ever been before, because of the greater loads and speeds,
but, at the same time, they are not impossible of solution.
Briefly stated, the wheel situation is this: Cast-iron wheels
under high-capacity cars are not giving altogether satisfactory
service, and with the keen competition and the severe tests
required, coupled with the unwillingness on the part of the rail-
roads to pay more for a higher quality wheel, the problem has
Metallurgical Notes and Comments 357
resolved itself into one of how good a wheel can be made for a
certain price. And this commercial aspect, as before said, has
caused both wheel makers and railroads to lose sight of the more
important mechanical problems. Mr. Griffin, speaking for the
wheel makers, admits that their chief aim is to make wheels as
cheaply as they can be made and stand the tests imposed by the
railroads, without regard to durability; and "the railroads, in
trying to save money on renewals, endeavor to turn back on the
makers as large a share as they can of the wheels which are con-
demned. Both parties are trying to save pennies by working at
cross purposes, when they might be saving dollars by working
together.
The cast-iron wheel has come in for much more than its
fair share of condemnation since the introduction of the high-
capacity car. Because it has failed under such cars more fre-
quently than under cars of lighter capacity, its enemies have
been quick to claim that it was unfit for such service, when, as
a matter of fact, almost any type of wheel would have failed
under similar conditions. The most frequent cause of failure
of wheels is the heating action of the brake shoes combined with
the rubbing of the flange against the rail. As far as vertical
strength is concerned, the cast-iron wheel apparently has an
ample reserve ; but under high-capacity cars the braking action
and the flange wear are both much more severe. Normally,
there is no flange wear on straight track, unless the truck fails
to return to its normal position after leaving a curve. If a car
has weak bolsters or badly designed center plates and side bear-
ings a truck may be held in a slued position, and the wheel flanges
grind on the inside of the rail mile after mile. Any wheel would
show flange wear under these conditions, and it is obviously
unfair to criticise the cast-iron wheel because it fails under such
conditions. The fault is not with the wheel.
The heating action of the brake shoes under heavy cars is
much more severe than under light cars, for two reasons: First,
because there is more braking pressure with practically the same
area for radiation of the heat; and, second, because, as a rule,
the high-capacity cars are braked more continuously — a larger
part of the time — than the lighter cars. In making up trains
the heaviest and strongest cars are usually put next to the en-
gine, and the light cars in the rear of the train. The trainmen
358 The Iron and Steel Magazine
then couple up the air brakes on the first ten or fifteen cars and
depend on the brakes of those cars to control the entire train.
This practice has been very general up to the time when the
statutory requirement of 50 per cent came into operation. On
mountain roads, where retaining valves are used, the front cars
in a train may have the brakes applied continuously for as long
as an hour, while the rear cars have only a few hand brakes set
up. This practice undoubtedly has much to do with the larger
percentage of wheel failures under high-capacity cars than under
cars of lighter capacity. We are not criticising the practice,
because experience has shown that in many situations probably
this is the best way to control a train ; we simply are pointing
out that it is not fairly the fault of the wheel when it fails under
such service.
We do not here discuss the relative merits of the difi'erent
designs of single and double plate wheels, with and without
brackets, which have been suggested as improvements over
existing designs. Experience only can determine whether or
not minor changes in the shape and location of the brackets and
plates will prevent troubles from cracks in the body of the wheel
due to uneven expansion, or whether they will assist in radiating
the heat from the tread to such an extent as to eliminate part of
the trouble from transverse and longitudinal cracks. The most
radical suggestion which has been made is to thicken the flange
and to enlist the cooperation of the maintenance of way depart-
ments to increase the clearance at frogs and guard rails so that
a thicker flange may be used. With a thicker flange more gray
iron can be used to reinforce the weakest point of the wheel
without sacrificing any of the chill. From experience on the
Southern Railway it would seem that this can be done with no
difficulty.
Viewing the whole matter impartially there would seem to
be no reason to doubt that some of the many expedients pro-
posed to improve the quality and increase the strength of cast-
iron wheels will eventually prove satisfactory; will remove all
ground for fear that the cast-iron wheel is unsafe and cannot be
made safe to run under the conditions of modern service. The
best care on the part of the founder and the use of high-grade
charcoal iron would eliminate many of the troubles now experi-
enced; and as regards this feature of the matter it is only a
Metallurgical Notes and Comments 359
question of whether the railroads prefer to pay a moderate price
for the better material and for the better foundry practice
adopted, or to pay much more for steel wheels, which appar-
ently have an excess of strength, but which have not yet shown
that they have sufficiently longer life to pay for the difference
in first cost. It is a significant fact that in their petition to the
Master Car Builders' Association, asking to have the guaranty
reduced, the wheel makers made no claims about being able to
produce a wheel to meet the present guaranty if the railroads
were willing to pay for it. This can hardly be taken as an ad-
mission that they cannot produce such a wheel, but it would have
made their position stronger had they offered a specification
and a guaranty for higher-priced wheels, and had shown that
they were prepared to furnish such wheels if the railroads wanted
them. Their attitude is thus a defensive one. and they have
not strengthened their position by taking it. Meanwhile, the
steel -wheel makers are taking every advantage of the situation
to secure favor, and by means which necessarily tend to create
distrust of the cast-iron wheel. The question is not yet by any
means settled, but nothing has occurred thus far to justify the
notion that the chilled- wheel makers must go out of business.
They know how to make better wheels, and they ought to know
how to induce the railroads to use better wheels. ^' Railroad
Gazette," July 14, 1905.
The Heroult Process in Sheffield. — The acquisition from the
Societe Electro-metallurgique Fran9aise, of Troyes, of the Brit-
ish patent rights in the Heroult process for the electrical produc-
tion of steel by the Sheffield Steelmakers (Lim.), of Sheffield, will,
if the expectations of the owners of the patents are fulfilled, mark
the beginning of a new era in steel manufacture in Sheffield.
Mr. P. R. Kuehnrich, the head of the Sheffield Steelmakers
(Lim.), has been interviewed by us, and has made the following
statements, which, it will be understood, are given entirely on
his own responsibilit}^ :
On inquiring what advantages were claimed for steel
smelted in the Heroult furnaces over that produced by other
methods, our representative was informed that the quality of
steel obtained by the electric process is not merely equal to the
highest grades of crucible carbon tool-steel, but is even better.
360 The Iron and Steel Magazine
especially as regards the uniformity of the material obtained.
In, say, a 4-ton charge of steel made in the Heroult furnace, the
whole of the ingots can be depended upon to agree in one uni-
form analysis, there being a complete absence of any tendency
to segregation.
Asked what is likely to be the immediate consequence of
the introduction of the Heroult process into Sheffield, Mr.
Kuehnrich expressed himself as follows: '' As a matter of fact
Sheffield can, if it wants to do so, immediately emancipate her-
self from Sweden, since in the Heroult furnace we can produce
from English scrap or ore iron purer than any coming from
Sweden. The analysis of some ingots of such iron which our
friends have made in the Heroult furnace reads as follows:
Sulphur 0.020
Phosphorus 0.005
Manganese 0.000
Carbon 0.0 10
Silicon 0.005
This may be termed chemically pure iron."
Our representative inquired whether Mr. Kuehnrich con-
sidered this '' Heroult " iron as good as the fine Swedish brands
which have been treasured in Sheffield for so many generations,
and have always been thought irreplaceable, and received the
following reply: ^' There are strong opinions — and I confess I
was myself formerly a holder of the theory — that there is some-
thing indefinable, but of unique character, in the choice brands
of Swedish and Styrian irons unattainable in any other material,
giving steel made therefrom that fine cutting and wearing quality
so much appreciated. It is true, Bessemer and Siemens-Martin
steels have been made with analyses practically matching those
of the steel made from Swedish and Styrian irons, but the
results did not bear out the anticipations of the analysis. The
Heroult process will not only affect the Swedish iron exports to
Sheffield, but will, I think, deal a blow at the Swedish Bessemer
and Siemens steelworks there as well. We claim to be able to
produce by the Heroult plant a special quality of steel which,
though as cheap in its production, is far superior to the Swedish
Siemens and Bessemer steel."
'' Do you expect that the Sheffield cutlery manufacturers
will take up your material? " '^ The introduction of our elec-
Metallurgical Notes and Comments 361
trie steel into Sheifield will tend to raise the standard of quality
of its cutlery, which during the last fifteen to twenty years has
deteriorated considerably. Generally speaking, Sheffield knives
do not now cut so well as formerly, since competition has forced
makers to employ cheaper qualities of steel than were used in
former days. ' Heroult ' electric steel is bound, I think, to help
to regain for Sheffield that prestige of ages which has undoubtedly
somewhat suffered."
" We claim that we can make high-class steel of a greater
uniformity than is possible by any other process. Therefore our
electric steel is the steel of the future for tools in general, rifle
barrels, swords, bayonets, heavy ordnance, armor-plates,
w res of high tensile and breaking strains, and almost endless
other purposes. But the matter does not end there. It will
set fresh standards for purposes where it has hitherto been
thought that a lower-grade steel was good enough. For instance,
for railway material, such as locomotive tires and axles, where
greater purity and uniformity of steel insures greater safety,
and for structural purposes, where lighter material will give the
•same strength as heavier sections of current type of steel mate-
rial. I foresee that railway companies, government departments
and corporations will alter the whole of their specifications,
which already have well-nigh given nightmares to the hapless
steel producers by ordinary methods." " The Iron Monger,"
July 22, 1905.
Some Notes on the Galvanizing Industry. — As compared
with the production of tin plate, galvanizing is comparatively
a new industry, having only been worked in Europe since the
early part of the last century and introduced into the United
States from England. The industry at the present time has
assumed large proportions, the value of the shipments of gal-
vanized roofing sheets amounting to over ;£i,ooo,ooo sterling
per annum, but the profits are comparatively small compared
with those made thirty years ago, when many large fortunes
were made.
The first method for coating iron with zinc was electrolytic,
that is, a solution of zinc such as zinc sulphate or chloride was
employed through which an electric current was passed, the
work to be zincked forming one pole and a sheet of zinc the other.
362 The Iron and Steel Magazine
The wet or electrolytic process was soon abandoned in favor of
the hot or molten process, as there were no efficient low-tension
dynamos which were essential for the economical working of such
a process, and the coating applied was found to be unreliable.
The hot or molten process consists of dipping the article into a
molten bath of zinc (at a temperature of about 800° F.) covered
with a suitable flux, such as sal ammoniac. From time to time
various improvements, relating to the mechanical details, have
been introduced into the working of the hot process, having for
their object the reduction of the amount of zinc consumed for a
given area of iron, but, unfortunately, such a reduction results
in a corresponding decrease in the durability of the galvanized
article. Sheets are passed through squeezing rolls as they leave
the bath; wire through asbestos rubbers or sand; nails and
similar articles are placed, when the zinc is still molten, in cen-
trifugal machines to remove as much zinc as possible. Another
improvement of some moment has been the introduction of close
annealing to reduce the amount of oxide formed, and the subse-
quent removal in acid. Sand-blasting, that is, projecting sand
at a few pounds pressure on the article to be freed from scale, has
been substituted for the wet acid process for some classes of
work. Serious attempts have been made on several occasions
to introduce lead galvanizing instead of zinc galvanizing, so as
to take advantage of the cheapness of the former metal, but lead-
coated articles are not found to be so durable as zinc coated, on
account of the lead being electro-negative to the iron instead of
positive as in the case of zinc. If the lead coating is penetrated
the iron is rapidly corroded ; in the event of the zinc coating being
perforated the zinc is slowly attacked and the iron preserved.
The electrical relation of zinc and lead to iron is readily demon-
strated by immersing in water pieces of zinc-galvanized iron and
lead-galvanized iron, the surfaces of both having been previously
penetrated. In a comparatively short time, in the first case,
zinc oxide will be noticed at the bottom of the vessel, and in the
other the water will rapidly redden and large patches of red
oxide of iron appear.
During the last ten years a number of electro-galvanizing
plants have been erected in almost every large shipbuilding yard
in England and on the Continent, as it has been found that zinc
can be very cheaply deposited with the modern efficient dynamo
Metallurgical Notes and Comments 363
of large amperage and small voltage. The process is employed
by the Admiralty for flashing or coating boiler tubes with a very
thin coating of zinc for the purpose of preventing corrosion and
pitting during the time of erection and assembly, and to enable
flaws to be detected in the tubes at an early stage, as a thin coat-
ing of zinc applied to the tubes is found to show up any inherent
defect very clearly. The process that has been adopted largely in
this country and by the German government, Messrs. Krupps,
Messrs. Cockerill and others on the Continent, is the regenera-
tive process, which is automatic in action. The zinc, as depos-
ited on the iron, is replaced by circulating the solution over zinc
dust (a product of the zinc distillation furnace) mixed with fine
coke so as to form a filter bed. Electro-galvanized iron is free
from the spangled or crystalline effect obtained by the hot pro-
cess, which is one of the reasons it has not been found suitable
for galvanizing sheets for roofing purposes, as certain foreign
markets attach great importance to this crystalline effect, al-
though many of our most enlightened engineers now specify
that galvanized iron shall not be spangled, as it is obtained at the
expense of the durability of the iron, tin being added in many
cases to increase the size and brilliancy of the spangles.
The latest development in the galvanizing industry is a
dry galvanizing process, which is an entirely new departure, as
is does not embody any of the essential features of the hot and
electric process. The new process, which is known as " Sherar-
dizing," consists in placing the iron article to be galvanized, with-
out any preparatory treatment beyond the removal of the matte
scale, in a closed receptacle filled with zinc dust, which is heated
to a temperature several hundred degrees below the melting
point of zinc, for such a period as will give the desired thickness
of zinc. The process for such articles as bolts, nuts, nails, small
castings, etc., is found to be more economical than hot galvaniz-
ing or electro-galvanizing, and the coating very durable, as a
true alloying or amalgamation of the zinc takes place with the
surface of the iron. The surfaces obtained by hot and cold
galvanizing and Sherardizing are different in each case, but they
can readily be distinguished by those conversant with the three
processes. In the case of hot galvanizing, the surface is span-
gled, or if not spangled, has the appearance of cast metal. In
the case of cold galvanizing, the surface is free from spangles,
364 The Iron and Steel Magazine
and has a matte or frosted surface, uniform if the work has been
well executed. Sherardizing is, again, distinctive from the two
former processes; the general appearance resembles more that
of cold galvanizing than hot galvanizing, but is more lustrous
and metallic, and is uniformly distributed over the whole sur-
face, which is not the case with the hot and cold galvanizing
processes. The hot and dry galvanizing processes are both con-
sumers of heat, and no doubt gas firing with Mond or a similar
gas will effect considerable economy over the present methods
of coke firing ; and the development of cheap electrical power on
a large scale should give a fresh impetus to the electro-galvaniz-
ing, which is entirely dependent on cheap electricity for its com-
mercial development, if it is to embrace any of the fields which
are at present entirely monopolized by the hot or molten process.
Great Britain was the first to establish the galvanizing
industry, which is now one of vast importance, and it is to be
hoped she will continue to take the lead by embracing those op-
portunities which occur from time to time of combining recent
improvements with cheap power and fuel, which will enable her
to retain many of her colonial and foreign markets, which are
now being contested so keenly by the Belgians, Germans and
Americans. Sherard Cowper-Coles in " Iron and Steel Trades.
Journal," July 8, 1905.
Nickel Steel Winding Rope Wire. — Mr. J. Devis has re-
cently carried out some experiments in order to test the possi-
bility of using nickel steel wires of low nickel content for hoisting^
purposes. He finds that wires with a high content of nickel are
not suitable for the manufacture of winding rope of high carryings
capacity, on account of their low strength and softness. Inas-
much as the capability of iron for taking up carbon decreases
considerably with increasing amounts of nickel present, and as
the high strength of steel wires is principally due to their higher
content of carbon, wires of a high carrying capacity are only
manufactured with a low content of nickel.
In the experiments special pains were taken to get true com-
parative results between the steel wires with or without nickel.
The first experiments were carried out with a nickel steel
wire, containing 5.74 per cent nickel and having a strength of
145,000 to 146,500 pounds per square inch. The results ob-
Metallurgical Notes and Comments 365
r
tained were as follows: Thickness of wire, average, 1.985 mm.;
tensile strength, 170,000 to 184,000 pounds per square inch;
number of torsions, 45.2; number of bendings, 16.2. Experi-
ments were also carried out in which the wires were subjected
to repeated shocks. A second series of experiments with the
same material was carried out to test the difference exerted upon
its mechanical properties after it had been exposed for eight
weeks to the action of steam above the exhaust of an engine and
had become covered with rust. The experiments showed that
the strength was not influenced to any large extent, but that
the bending capacity was reduced by 41,4 per cent, and the tor-
sional capacity had suffered the still greater reduction of 49.7
per cent. The presence of nickel, amounting to 5.74 per cent,
therefore, exercised no influence to prevent the rusting.
Comparative experiments, carried out with winding rope
wires of the same steel material but without the addition of
nickel, which had nearly the same strength as the wire contain-
ing nickel, namely, 170,000 to 184,000 pounds per square inch,
showed that these wires possessed the same bending capacity as
the nickel wires tested previously, but had a smaller torsional
capacity. Experiments were also conducted with these wires
after they had been rusted in the same manner as above. They
showed that the strength decreased somewhat, while the bending fj
capacity decreased 33.8 per cent, and the torsional capacity
24.7 per cent. Other experiments were carried out with nickel
steel wires containing 1.89 per cent nickel, which had a strength
of 256,000 to 284,000 pounds per square inch, and with ordinary
steel wires of about the same strength. The experiments
showed that a favorable influence of the nickel present on the
mechanical properties of the wire could not be ascertained.
In order to find out whether a larger amount of nickel in
the wire exerts a favorable influence on its properties, nickel
steel wires were tested, containing 6.28 per cent nickel, which
had a carrying capacity of 256,000 to 284,000 pounds, and com-
pared with the results obtained from ordinary steel wires of the
same material, as to their resistance to the influence of rust and
their mechanical properties, as well as to their behavior when
subjected to bending, torsion and shocks. It was found that
in this case also a rust-preventing property of the 6.28 per cent
nickel in the wire could not be established.
366 The Iron and Steel Magazine
In reviewing the experiments, Mr. Devis thinks that at
the present time nickel steel cannot come into consideration as
a material for winding ropes. ^^ Iron and Coal Trades Review,"
August 4, 1905.
British '^^^ second annual convention of the British
Foundrymen's Foundrymen's Association was held in
Glasgow on August 7, 8 and 9. The mem-
bership, which at the last meeting was 83, now stands at 121.
It was reported that the council had made arrangements
for the publication of a journal in which would appear the
papers read as well as the other transactions of the society.
Messrs. Robert Buchanan, Herbert Pilkington and F. W. Finch
were re-elected respectively president, vice-president and secre-
tary.
Mr. Buchanan delivered his presidential address and the
following papers were read and discussed:
'' Technical Education and the Foundry, " by Professor
Sexton.
'^ Cast Iron," by Herbert Pilkington.
'' Molding Sands and Fire Clays," by Percy Longmuir.
" The Microscope and Pig Iron," by A. Campion.
^' Profitable Foundring," by John G. Stewart.
Moisture and Furnace Results. — At the Liege Metallur-
gical Congress, M. Divary, of Schneider & Co., Creusot, France,
gave some interesting information on the occasion of the discus-
sion of Mr. Gayley's dry-air process. The observation was made
for many years at Creusot that consumption of coke increased
and the output of the furnace decreased in summer as compared
with winter, and in 1901 records were kept, which in August,
1902, led to a first study . The result of this study was that an
economy of 50 to 60 kilograms of coke per ton of iron seemed in
sight if the furnace was blown with air saturated at 0° C.
in place of hot and humid air during the summer months.
These researches were continued, and since the difference in
consumption of coke observed during one year seemed larger
than those justified by calculation it was determined to carry
out the system of drying the air by freezing. This was sub-
mitted in September, 1903, to several designers of ice machines.
Metallurgical Notes and Comments 367
After their figures had been received other questions claimed the
attention of Schneider S: Co., so that the program was not imme-
diately carried out, but instructions were given to the blast-
furnace department to observe daily the humidity in the air
and the corresponding fuel consumption under conditions guar-
anteeing accuracy. Three times per day the moisture was
observed in the blowing room.
During the year 1904 the charge remained practically the
same, the two furnaces, Nos. 2 and 4, running on basic-Besse-
mer pig. The results are shown in the following table, the first
column showing the average amount of moisture per cubic meter
of air blown in during the month, the second column the increase
in the consumption of coke as compared with that of the driest
month, and the third column showing the average production
per furnace for twenty -four hours:
Moisture Increase in Coke Average Furnace
Month in Blast Consumption Product
Gr. Kg. Tons
January 6.3 o 90.5
February 6.6 10 88.7
March.. 7.6 13 87.2
April 7.8 47 81.2
May 10. o 56 78.3
June II. 7 103 75.7
July • 130 133 70.0
August 12.0 90 76.6
September 9.3 55 90.0
October. 8.0 28 86.0
November 7.6 25 86.5
December 7.0 35 88.5
At Creusot the speed of the engines is not lowered in winter,
and the blast is always heated to the same temperature, summer
and winter, being an average of 752° C.
The figures agree very closely with those of Mr. Gayley
and hold out undoubtedly the results to be obtained from drying
blast-furnace blast. " Iron Age," August 17, 1905.
Production of Manganese Ores in 1904. — California, Utah
and Virginia were the only states that produced manganese ore
in 1904. The total production for the year amounted to 3,146
gross tons, valued at $29,466, or $9.37 a ton. This is 321 gross
tons, or II per cent, more than the quantity reported in 1903,
which was 2,825 gross tons. Of the total production in 1904,
3,054 gross tons, or 97 per cent, came from Virginia, 60 gross
368 The Iron and Steel Magazine
tons, or 2 per cent, came from California and 32 gross tons, or
I per cent, came from Utah.
In addition to the true manganese ores considerable quan-
tities of manganiferous iron ore are produced in Arkansas, Colo-
rado and in the Lake Superior region. This amounted in 1904
to 383,246 gross tons, which had a reported value at the mines
of $691,477. In Arkansas 600 tons of ore of this class, carrying
28 per cent of manganese and 10 to 14 per cent of iron, were
mined and used in the manufacture of pig iron. The Colorado
ores are utilized primarily as flux by the precious metal smelters,
the remainder being employed in the manufacture of spiegeleisen.
In the Lake Superior region quantities of iron ore are mined
which analyze from a fraction of i per cent up to 20 per cent of
manganese.
In mining silver ores in Colorado a considerable quantity
of ore is obtained which contains insufficient percentages of the
precious metal to make it valuable on that account, but which
is used as a flux by the smelters. This ore is considered an iron
ore and is included in the report on that mineral. In 1904 the
production of manganiferous silver iron ores in the United
States amounted to 105,278 gross tons, valued at $348,132.
This is considerably less than the production of 1903, which
amounted to 179,205 tons, worth $649,727.
A by-product in the manufacture of zinc from ores mined in
northern New Jersey, containing iron and manganese, is utilized
in the production of spiegeleisen. In 1904, 68,189 gross tons
of this class of ore were obtained.
The total quantity of manganese ore, manganiferous iron
ore, argentiferous manganiferous ore and zinc residuum pro-
duced in the United States in the calendar year 1904 amounted
"to 559.859 gi'oss tons.
The above-mentioned facts are taken from a recent report
made by Mr. John Birkinbine for the United States Geological
Survey. " Bulletin," American Iron and Steel Association,
August 15, 1905.
The Wonderful Lake Superior Region. — We compile below,
from the report of John Birkinbine to the United States Geo-
logical Survey on the production of iron ore in 1904, a table
showing the relative prominence of the Lake Superior region as
MctaUnrgical Notes aiid Comments
369
an iron ore producer in the last three years as compared with the
remainder of the country. The loss in production in 1904 will
doubtless be fully balanced by an increased production in 1905.
States — Gross Tons
1902
1903
1904
^Knnesota
15.137.650
11,135.215
783,996
15.371.396
10,600,330
675.053
12,728,835
7,089,887
483,745
Michigan
Wisconsin
Total
27,056,861
8,497,274
26,646,779
8,372,529
20,302,197
7,342,133
All other states
Grand total
35.554,135
35,019,308
27,644,330
The total value of Lake Superior iron ore at the mines in
1902, as given by Mr. Birkinbine, was $52,485,951; in 1903,
$53,858,635; and in 1904, $32,934,873. Of this value much the
larger part was, of course, paid to labor. The shrinkage in
production and value in the reactionary year 1904, as compared
with the two preceding years, shows how seriously the iron ore
interests of the Lake Superior region are affected by a depression
in the American iron trade.
The above figures make plain a truth that needs to be
generally recognized, although it is self-evident. A brisk de-
mand for the raw materials of our iron and steel industries, iron
ore, coke, coal, limestone, etc., cannot be maintained if the
demand for the finished products declines from any cause what-
ever. '' Bulletin," American Iron and Steel Association, August
15. 1905-
Xhe The accompanying illustration shows a machine
Brinell Ball constructed in Sweden for determining the hard-
ness of metals by the Brinell method, which consists
in forcing a steel ball into the metal under a known pressure and
in measuring the spherical depression produced thereby. By
dividing this depression by the corresponding load the " factor
of hardness " is obtained. From this factor of hardness the
tenacity of the metal may also be calculated with sufficient
accuracy for most practical purposes, thus doing away with the
costly preparation of test pieces for the ordinary tensile test.
The ball test may, moreover, be applied to finished implements.
370
The Iron and Steel Magazine
such as rails, projectiles, armor plates, machiner}^ parts, etc.
Its usefulness in ascertaining the effect of hardening and
tempering will also be obvious. The hydraulic machine illus-
trated here for conducting the ball test attracted much atten-
tion at the Liege Exposition. It is said to be very practical, of
easy manipulation and absolutely accurate. The machine is
constructed by the Alpha Company (Aktiebolaget Alpha), of
Stockholm, Sweden.
REVIEW OF THE IRON AND STEEL MARKET
September has been a remarkable month in the American
iron trade. There has been an improvement in every individual
line. The demand for steel for the different finishing lines has
become so great that scarcely any crude steel can be obtained
in the open market. This demand has in turn been reflected
in pig iron, but not to so great an extent, as demand for forge
and foundry grades has not been abnormally heavy.
During October substantially the entire steel producing
capacity of the country v^ill be operated to the full limit, even
including the old Columbus Bessemer steel works, which are run
only in times of extreme demand. The pressure is not uniformly
distributed among the various, finishing lines. In rails, shapes
and plates the demand is far ahead of capacity. In merchant
steel bars it is fully up to capacity. In wire products, sheets,
merchant pipe and tin plates it is scarcely up to total finishing
capacity, the deficiency being somewhat in the order in which
these products have been named. It is always the case with the
trade that the total finishing capacity, in all lines, is beyond the
steel producing capacity. It would be simply impossible to
supply steel for operating all finishing mills to full capacity, and
the trade is already on the verge of an absolute famine in crude
steel .
Pig Iron. — It can be stated as a general proposition that
all the blast furnaces of the country are in operation except
such as are off for absolutely necessary repairs and such as it is
not economical to operate even with the present moderately
high market prices. Production is at a slightly higher rate than
at any time since July i, but is under the rate maintained for
a short time in the latter part of the first half. At that time
furnaces were in better than normal condition, owing to much
relining and repair work having been done in the lull last year.
Sales of all grades have been fairly heavy. The market has
advanced an average of about 50 cents during September.
Furnaces in general are well sold up for several months ahead.
Z7^
372 The Iron and Steel Magazine
A number of steel works producing pig iron themselves have
been good buyers of Bessemer and basic. The United States
Steel Corporation has resumed making monthly purchases of
pig iron, taking 35,000 tons of Bessemer for September delivery
and a larger quantity for October. We now quote prices as
follows: F.o.b. valley furnace: Bessemer and basic, $15.00;
No. 2 foundry, $14.75 "to $15.00; gray forge, $13.90 to $14.00;
low phosphorus, $20.75. Delivered Pittsburg: Bessemer and
basic, $15.85; No. 2 foundry, $15.60 to $15.85; gray forge,
$14.75 "to $14.85. F.o.b. Birmingham: No. 2 foundry, $12.00;
gray forge, $10.75, ^^^ nearby delivery, most southern furnaces
asking 50 cents advance for next year's delivery. Delivered
Philadelphia: No. 2 X foundry, $16.50 to $16.75; standard gray
forge, $14.75 to $15.25. Delivered Chicago: northern No. 2
foundrv, $16.25 to $16.50; malleable Bessemer, $17.00 to $17.25.
Freight: Birmingham to Pittsburg, $4.35; to Cincinnati, $2.75;
to Chicago, $3.65; to Philadelphia by water, $3.50; by all-rail,
$4.00.
Steel. — After having been almost the only seller of crude
steel during August, the Carnegie Steel Company about the
first of September withdrew entirely from the market, and no
seller of consequence appeared to take its place. While con-
sumers in general were well covered, inquiries aggregating
more than 30,000 tons have been turned down. The only sale
of any magnitude was one of 10,000 tons of Bessemer billets
from a producer to another producer, making only open-hearth
steel. Such transactions are not market indications, as steel
producers make special efforts to accommodate each other.
We quote Bessemer billets largely nominal at $25.00 and open-
hearth at $26.00, these prices representing an advance of $1.00
a ton during the month. There are no open quotations on
sheet bars, but the settlement price on long-term contracts,
subject to monthly price adjustment, is $26.00 for October
deliveries, a large tonnage being controlled by such contracts.
Forging billets in small lots are bringing as high as $28.00. Rods
are quoted at $31.00 to $32.00. All prices are f.o.b. Pittsburg.
Rails. — Since the last week in August rail mills have been
booking tonnage for 1906, at $28.00, f.o.b. mill, which has been
the price since the advance made in the spring of 1901. Fully
1,500,000 tons have been booked, and a considerable tonnage,
Review of the Iron and Steel Market 373-
perhaps 250.000 tons, will have to go over from this year, so that
more than half the record production for any year has already
been sold, with every indication that 1906 will break records.
Production this year will be much larger than early estimates,
as in the second half all possible capacity is engaged on rails,
while in the first half a considerable tonnage of billets and sheet
bars was being produced by the convertible mills.
Shapes. — On August 3 1 all structural shapes were advanced
$2.00 a ton, making official prices the highest since 1899, as
follows: Beams and channels, 15-inch and under, angles 2x3
to 6 X 6 inclusive, and zees, 1.70 cents per pound; tees, 3 -inch
and larger, 1.75 cents; beams and channels over 15-inch, 1.80
cents, all f.o.b. Pittsburg, carload and larger lots. The large
structural mills are filled with actual specifications which it will
take from five to six months to roll, and deliveries for this year
can be obtained only from stocks which are much depleted, and
from a few small mills which do not sell so far ahead. A large
tonnage is being booked for 1906 delivery, chiefly for definite
undertakings, a few running contracts being made with fabri-
cators and dealers.
Plates. — During September orders have been placed by
various railroads for about 50,000 steel and steel underframed
cars, engaging the capacity of car builders until about July i
next. Plates for these cars have been placed as the car orders
came in, and there has also been good buying for shipbuilding
and other purposes for next year. The plate mills are, on an.
average, sold up for four or five months ahead, and early de-
liveries are hard to get. A general advance in plates is likely
to be made during October. The mills have already advanced
plates 14 inches wide and narrower by $2.00 a ton, placing
them on the same basis as wider plates. We now quote the
regular mill price at 1.60 cents for tank quality, quarter- inch
and heavier, 6\ to 100 inches wide inclusive, with the usual
advances for other sizes and grades.
Merchant Bars. — Contracting for future delivery con-
tinues very heavy, and specifications are in hand for several
months ahead. Hoops have been advanced $2.00 a ton to 1.75
cents, base, with full hoop extras. Merchant steel bars con-
tinue at 1.50 cents. An advance in both hoops and steel bars
will probably be made during October. Common iron bars.
374 The Iron and Steel Magazine
have been steadily advancing, and the market is now 1.70 cents,
f.o.b. Youngstown, 1.75 cents, Pittsburg, and 1.65 cents,
Chicago.
Sheets. — A very large tonnage of sheets has been booked,
and this, with the steadily advancing prices of steel, should
have resulted in a fair advance in sheets before this, but so far
no more can be reported than that extreme concessions have
been withdrawn in the past fortnight. So far as ordinary car-
load business is concerned, the market is lower than it was a
month or two months ago. We quote on ordinary carloads for
mill shipment, with desirable specifications, 2.25 to 2.30 cents
on black and 3.30^0 3.35 cents on galvanized. No. 28 gauge.
On very large lots these prices could be shaded $1.00 a ton, but
not more. Some of the independent mills are filled for the pres-
ent, and will not do these prices. The leading interest has
booked a large tonnage and is now operating more than 85 per
cent of its sheet capacity. Corrugated roofing is $1.65 to $1.75
per square for painted and $2.85 to $2.95 for galvanized. No.
28 gauge. The special rebate of 15 cents a box on tin plate, in
addition to the regular rebate of 5 cents a box from the official
price of $3.55 for 100-pound cokes was made to expire October 5,
but as the market is but slightly improved no great change in
net prices is to be expected, although the form of quotation may
be changed.
Wire Products. — New official prices have been put out,
based on $1.75 for nails in carload and larger lots to jobbers.
These prices represent a reduction of $1.00 a ton in official quo-
tations, but former quotations were being regularly shaded
$2.00 a ton or more, so that the market is really higher. Ton-
nage has been good and wire mills are operating to almost full
capacity. Production was restricted during July and August
and stocks are not large.
Scrap. — Prices have advanced further, and dealers are in
expectation of a further advance when wintry weather sets in,
so that they are accumulating considerable scrap, and it is be-
lieved paying higher prices in some instances than could be real-
ized on prompt sales to consumers. We quote heavy melting
stock at about $16.00, delivered Pittsburg. Cast borings are
firm at our former quotation of $9.50 to $10.00. vSheet scrap is
bringing $14.25.
STATISTICS
Half Yearly Production of Pig Iron.* — The following table
gives the production of pig iron in the United States in half-
yearly periods from 1883 to 1905, in gross tons:
Years — Gross Tons
First Half
Second Half
Total
18S3
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
2,352,019
2,024,126
1.920,371
2,637,687
3.049.295
3,020,092
3,661,603
4.560,513
3.368,107
4,769,683
4,562,918
2,717.983
4,087,558
4.976,236
4,403,476
5.869,703
6,289,167
7,642,569
7.674,613
8,808,574
9.707,367
8,173.438
11,163,175
2,243,491
2,073,742
2,124,155
3.045.642
3.367.853
3,469,646
3.942,039
4,642,190
4,911.763
4,387.317
2,561,584
3,939,405
5.358,750
3,646,891
5,249,204
5,904,231
7.331.536
6,146,673
8,203,741
9.012,733
8,301,885
8,323.595
4,595.510
4,097,868
4,044,526
5,683,329
6,417,148
6,489,738
7,603,642
9,202,703
8,279,870
9,157,00a
7,124,502
6,657,388
9.446,308
8,623,127
9,652,68a
11.773.934
13,620,703
13,789,24^
15,878,354
17,821,307
18,009,252-
16,497.033
Total Production of all Kinds of Pig Iron according to Fuel
Used.*
Blast Furnaces
Gross Tons of 2,240 lbs (Includes
In
blast
Dec.
31,
1904
June 30, 1905
Spiegeleisen, Ferro-Manganese, etc.)
First half
of 1904
Second half
of 1904
In
Out
Total
First half
of 1905
Bituminous
Anthracite
Charcoal
206
38
17
229
42
23
82
311
65
56
7.337.279
622,803
213.356
7.594,085
605,337
124,173
10,162,488
830,17s
170, 51^
Total . . . .'
261
294
138
432
8,173.438
8,323.595
".163,175
* " Bulletin," American Iron and Steel Association.
375
376 The Iron and Steel Magazine.
Coke Production in 1904. — Mr. Edward W. Parker, of
the Division of Mining and Mineral Resources of the United
States Geological Survey, reports a coke production during the
calendar year 1904 of 23,621,520 net tons, valued at $46,026,183,
or $1,948 per net ton. There were 83,499 ovens built and 4,430
building. To this production Pennsylvania contributed 14,861,-
064 tons, or over 62 per cent. The productions of coke from
bv-product ovens amounted to 2,608,229 tons, or a little over
1 1 per cent of the total production. Of the completed by-product
ovens in 1904, there were 895 Semet-Solvay, 1,795 Otto-Hoff-
man and Schniewind, 56 Newton-Chambers, 66 Wilcox and 98
Rothberg. The average production of the by-product ovens
was 896 tons each, while the average production of the beehive
ovens was 283.5 "tons.
The British Iron Trade in 1904. — The following statistics
are from the report of Mr. J. Stephen Jeans, secretary of the
British Iron Trade Association:
Iron Ore. — The total production of iron ore in 1904
amounted to 13,774,282 gross tons, against 13,715,645 tons in
1903. The total imports of iron ore in 1904 amounted to
6,100,756 tons, against 6,314,162 tons in 1903.
Bessemer Steel Rails. — The production of Bessemer steel
rails in 1904 amounted to 916,374 tons, as compared with 1,061,-
441 tons in 1903 and 903,216 tons in 1902.
Puddled Bars. — The production of puddled bars in 1904
amounted to 936,228 tons, against 950,393 tons in 1903. The
number of completed puddling furnaces in 1904 was 1,470, of
which 1,166 were active in the year and 304 were idle.
Tinplates. — The production of tinplates in 1904 is esti-
mated at 12,500,000 boxes, of which 8,000,000 boxes were for
export and 4,500,000 boxes were for home consumption.
Iron and Steel Shipbuilding. — On December 31, 1904, the
tonnage of iron and steel vessels under construction amounted
to 1,049,860 tons, against 898,478 tons on December 31, 1903.
RECENT PUBLICATIONS
Lcs Aciers Speciaux, Volume II, by Leon Guillet. 132
9 X II -in. pages; numerous illustrations, mostly photo-micro-
graphs. Paper covers. Vve. Ch. Dunod. Paris. 1905. Price,
10 francs. — This second installment of Mr. Guillet's important
work dealing with special steels includes his papers on chromium,
tungsten, molybdenum, tin, titanium, vanadium, aluminum
and cobalt steel, which were originally published in the " Revue
de Metallurgie " in 1904 and 1905. It will be remembered
that the first volume, which was reviewed in The Iron and
Steel Magazine for October, 1904, dealt with silicon, man-
ganese and nickel steel. Like the first volume, the present one
includes a preface by Prof. H. Le Chatelier. In his well-
known researches the author considered especially the structure,
tenacity, elastic limit, ductility, hardness (by Brinell's test) and
brittleness (by the testing of nicked bars) of the steels which he
studied. His experiments were conducted in a very systematic
manner and included samples containing from 0.20 to 0.80 per
cent of carbon and increasing proportions of the special element,
in some cases up to 30 per cent. The value of Mr. Guillet's
work needs no demonstration. Professor Le Chatelier does
not hesitate to compare it to the classical researches of Sir
Lowthian Bell on blast-furnace phenomena and to those of
Hadfield on special steels. Metallurgists and engineers inter-
ested in these important steels are also under great obligation to
the Steel Works of Imphy (France) and to the Dion et Bouton
Company for the generosity with which the first-named provided
the needed steels, and the latter the laboratory facilities, for
their treatment and the study of their properties.
Procedes Metallurgiques et Etti.de des Metaux, by U. Le
Verrier. 403 6\ X lo-in. pages; illustrated. Paper covers.
Gauthier-Villars. Paris. 1905. Price, $3.60. — This is a recent
volume added to the publishers, " Encyclopedic Industrielle ""
2>n
37^ The Iron and Steel Magazine
series. The first part of this book deals with ores and metal-
lurgical processes and contains chapters on sampling, dressing,
drying, calcination and roasting, classification and description
of furnaces, extracting methods, reduction, treatment of sul-
phide ores, distillation, electric processes, wet methods, smelting,
refining, thermo-chemistry, metallurgical appliances; the sec-
ond part is devoted to the physical testing of metals and to the
influence of heat; the third part to metallography and the
fourth part to alloys.
It will be seen that the subject covered is a very wide one
and the author's treatment is necessarily general in its character.
The usefulness of books dealing with the essential chemical,
physical and mechanical features of metallurgy is not to be
denied. The information conveyed by such treatises forms a
safe foundation upon which to rest specialized studies; it con-
stitutes the introductory knowledge which shotild be acquired
before venturing into any special metallurgical field. Professor
Le Verrier's book is a notable addition to metallurgical litera-
ture. We shall, however, venture the criticism that some of
the photo-micrographs reproduced to illustrate types of struc-
ture do not seem to have been very well selected, while the poor
quality of the paper used render some of them quite unintelli-
gible, such being especially the case with the illustration on page
370-
Aids to the Analysis and Assay of Ores, Metals, Fuels, etc.,
t>y J- James Morgan. 105 4 X 6|-in pages. Bailliere, Tindall
■& Cox. London. 1902. Price, 2s. 6d. — In this little book,
which is for the first time brought to our attention, the author
gives brief description of standard methods for the analysis of
ores, metals, fuels, etc. In many instances only one method is
given, while more than two methods are seldom outlined for
any one determination, but those described appear to have
been carefully selected. The special value of the book for
ready reference lies in its condensed form, no fewer than 188
chemical determinations, covering a very wide metallurgical field,
being described in less than 100 small pages.
The Journal of the Iron and Steel Institute, Vol. LXVII
(No. I, 1905). Edited by Bennett H. Brough, secretary. 865
Recent Publications 379
4^ X 8-in. pages; illustrated. E. & F. N. Spon. London. 1905. —
This volume, which, if we are not mistaken, contains more pages
than anv of those previously published, includes the minutes
of the annual general meeting held in London in May, 1905, and
the usual number of carefully compiled " Notes on the Progress
of the Home and Foreign Iron and Steel Industries." A recent
photograph of Right Hon. Sir Bernhard Samuelson, Bart.,
F. R. S., past president of the Institute, is reproduced as a
frontispiece. The minutes of the proceedings include R. H.
Hadfield's presidential address and the following papers with
their discussions:
" The Continuous Steel Process in Fixed Furnaces." By
S. Surzycki.
" Recent Developments of the Bertrand-Thiel Process in
the Manufacture of Steel." By J. H. Darby and G. Hatton.
" Experiments Relating to the Effect of Mechanical and
Other Properties of Iron and Its Alloys Produced by Liquid
Air Temperatures." By R. A. Hadfield.
" The Applications of Dry-Air Blast " (supplementary
paper). By James Gayley.
" The Cleaning of Blast-Fumace Gas." By Axel Sahlin.
" Experiments on the Fusibility of Blast-Fumace Slags. '^
By O. Boudouard.
" Notes on the Failure of an Iron Plate through Fatigue."
By S. A. Houghton.
" Accidents Due to the Asphyxiation of B last-Furnace
Workmen." By B. H. Thwaite.
" Sulphur in Coke, and Its Behaviour in the Blast Furnace."
By F. Wuest and P. Wolff.
" The Types of Structure and the Critical Ranges on Heat-
ing and Cooling of High-Speed Tool Steels under Varying Ther-
mal Treatments." By H. C. H. Carpenter.
" Magnetic and Electric Properties of Various Kinds of
Sheet Steel and Steel Castings." By Gunnar Dillner and A. F.
Enstrom. (Abstract.)
" Effects Caused by the Reversal of Stresses in Steel." By
G. C. Gardner. (Abstract.)
" Troostite." By F. Rogers. (Abstract.)
" Heat Treatment and Fatigue of Steel." By F. Rogers.
(Abstract.)
380 The Iron and Steel Magazine
Elementary Microscopy, by F. Shillington Scales. 179
4 X 6-in. pages; 78 illustrations. Bailliere, Tindall & Cox.
London. 1905. Price, 35. — This is essentially a handbook for
beginners, the author confining himself to a clear and concise
description of the optical principles involved in the construction
of microscopes and of the manipulation required in their use.
The book contains illustrations of the standard types of the
best English microscope makers. The book is well printed on
good paper and the illustrations are satisfactory.
The New Knowledge, by Robert Kennedy Duncan, pro-
fessor of chemistry in Washington and Jefferson College. 263
4^ X 8-in. pages; illustrated. A. S. Barnes & Co. New York.
1905. Price, $2.00. — In this exceedingly interesting book the
author attempts to present in a popular and yet strictly scientific
manner, the recent contributions of such eminent physicists
and chemists as Becquerel, the Caries, Ramsa}^ Crookes and
others. The subject is sub-divided into seven parts, dealing re-
spectively with Current Opinions, The Periodic Law, Gaseous
Ions, Natural Radio- Activity, The Resolution of the Atom,
Inorganic Evolution and New Knowledge and Old Problems,
each part being in turn subdivided into several chapters. The
author's style is clear and forceful and his book altogether
delightful and singularly instructive reading.
A Technological and Scientific Dictionary. Parts I to IX
inclusive. Edited by G. F. Goodchild and C. F. Tweney. Each
part has 64 6 X lo-in. pages and is illustrated. George Newnes,
Ltd. London. Price, each part, i5. net. — This dictionary will
be completed in fifteen parts and will then contain the definition
of some 16,000 words used in pure and applied science, such
as chemistry, physics, engineering, textile industries, architec-
ture, etc. The dictionary includes special articles written by
■experts on important subjects, such as iron, steel, gas manu-
facture, boilers, steam engines, etc. It should be welcome by
workers in pure as well as in applied science.
Statistical Abstracts of the United States, IQ04. Prepared
by the Bureau of Statistics, under the direction of the Secretary
Recent Publications 381
of Commerce and Labor. 659 6 X 9-in. pages. Government
Printing Office. Washington. 1905. — This is the twenty-sev-
enth annual volume of statistical abstracts prepared by the
Bureau of Statistics, and it contains the usual amount of valu-
able information referring to most of the natural and manu-
factured products of the United States during the last calendar
year and many financial tables of the greatest value to those
interested in the business condition of the country at large.
The volume contains 203 statistical tables.
Society for the Promotion of Engineering Education, Vol.
XII. Edited by C. F. Allen, F. W. McNair and M. S. Ketchum,
committee. 254 6 X 8^-in. pages. Engineering News Pub-
lishing Company. New York. 1905. Price, $2.50. — This vol-
ume contains the proceedings of the twelfth annual meeting
of the society, which was held at St. Louis September 1-3, 1904.
International Catalogue of Scientific Literature . Seventeen
volumes, 8vo. Published for the International Council by the
Royal Society, London. Harrison & Son. Price, ;^i8. —
This " International Catalogue of Scientific Literature," com-
mencing with the literature of the year 1901, is an outgrowth
of the " Catalogue of Scientific Papers Relating to the Scientific
Literature of the Nineteenth Century," published by the Royal
Society of London.
The possibility of preparing a complete index of current
scientific literature by international cooperation was first
taken into consideration by the Royal Society about the year
1893. The society sought the opinion of a very large number
of representative bodies and individuals abroad; and, as the
replies were almost uniformly in favor of the work being under-
taken by international cooperation, an International Conference
of Delegates appointed by various governments took place in
London in 1896. It was unanimously resolved that it was
•desirable to compile and publish, by means of an international
organization, a complete catalogue of scientific literature, ar-
ranged according both to subject matter and to authors' names,
in which regard should be had, in the first instance, to the re-
quirements of scientific investigators, so that these might find
out, with a minimum of trouble, what had been published on
382 The Iron and Steel Magazine
any particular subject of inquiry. It was agreed that the
material should, as far as possible, be collected in the various
countries by local organizations established for the purpose,
and that the final editing and publication of the catalogue
should be intrusted to a Central International Bureau acting
under the direction of an International Council, and it was
agreed to establish the Central Bureau in London.
At a subsequent meeting it was decided that the Ro3^al
Society be requested to organize the Central Bureau, and to do
all necessary work, so that the preparation of the catalogue
might be commenced in 1901.
The supreme control over the Catalogue is vested in an
International Convention. Such a convention is to be held in
London in 1905, in 19 10, and every tenth year afterwards, to
reconsider and, if necessary, to revise the regulations for carry-
ing out the work of the Catalogue; but the approved schedules
are not to be altered during the first period of five years. In
the interval between two successive meetings of the convention,
the administration of the Catalogue is vested in an International
Council, the members of which are to be appointed by the
Regional Bureaus.
The first meeting of the International Council was held in
London on December 12, 1900, when it was decided to com-
mence the preparation of the Catalogue from January i, 1901.
At this meeting an executive committee was appointed, con-
sisting of the delegates of the Royal Society and representatives
of the four largest subscribers, — the United States of America,
Germany, France and Italy.
The materials out of which the Catalogue is formed are to be
furnished by Regional Bureaus.
The branches of science to be included in the Catalogue are
the seventeen following: A- — Mathematics; B — Mechanics;
C — Physics; D — Chemistry; E — Astronomy; F — Meteor-
ology (including Terrestial ■ Magnetism) ; G — Mineralogy (in-
cluding Petrology and Crystallography); H — Geology; J —
Geography (Mathematical and Physical) ; K — Palaeontology ;
L — General Biology ; M — Botany ; N — Zoology ; O —
Human Anatomy; P — Physical Anthropology; Q — Physi-
ology (including experimental Psychology, Pharmacology and
Experimental Pathology) ; R — Bacteriology.
Recent Publications 383
Each complete annual issue of the Catalogue will thus
consist of seventeen volumes. The price at which this set will
be sold to the public is ;^i8. Individual volumes will be sold at
prices varying with their size, from about 10 to 355,
Each volume consists of three parts: (a) Schedules and
indexes in English, French, German and Italian, (6) an authors'
catalogue and (c) a subject catalogue.
The gigantic size of this undertaking will be readily appre-
ciated as well as its undoubted value to scientific workers, but
the high price at which it is apparently necessary to sell the
volumes is much to be regretted, for it will be found prohibitive
for a large number of interested persons. The Smithsonian
Institution at Washington has undertaken the work of indexing
the literature of the United States, and those desiring to become
subscribers to the Catalogue should apply to that institution.
BOOKS RECEIVED
The following books have been received and will be reviewed in an
early issue of The Iron and Steel Magazine.
Coke. A Treatise on the Manufacture of Coke and Other Prepared
Fuels and the Saving of By-products, by John Fulton. 476 6 X 9-in.
pages; illustrated. International Textbook Company. Scranton. 1905.
Price, $5.00.
Mechanics of Materials, by Mansfield Merriman. Tenth edition,
rewritten and enlarged. 507 6 X 9-in. pages; illustrated. John Wiley &
Sons. New York. 1905. Price, $5.00.
Machine Shop Tools and Methods, by W. S. Leonard. 554 6 X 9-in.
pages; nearly 700 illustrations. John Wiley & Sons. New York. 1905.
Price, $4.00.
Laboratory Notes on Practical Metallurgy, by Walter MacFarlane.
140 5 X 7 in. pages; illustrated. Longmans, Green & Co. New York.
1905.
Laboratory Chemistry, by Richard B. Moore. 195 5 X 7 -in. pages;
illustrated. J. B. Lippincott & Co. Philadelphia. 1904. Price, 75 cents.
Metallurgy of Cast Iron, by Thomas D. West. Ninth edition. 677
5 X 7-in. pages; illustrated. The Cleveland Printing and Publishing
Company. Cleveland, Ohio. 1904. Price, $3.00.
The Copper Handbook, by Horace J. Stevens, Volume V (1904).
882 6 X S^-in. pages; illustrated. Houghton, Mich. 1905. Price in
buckram binding, $5.00; in full morocco, $7.50.
The Crystallization of Iron and Steel, by J. W. Mellor. 144 4 X 7^-in.
pages; illu.strated. Longmans, Green & Co. New York. 1905. Price,
$1.60.
Friction and Lubrication, by William M. Davis. Second edition.
The Lubrication Publishing Company. Pittsburg, Pa.
PATENTS
RELATING TO THE METALLURGY OF IRON AND STEEL
UNITED STATES
795,139. Blast Furnace. — Nelson M. Langdon, Mancelona, Mich.
795,193. Treatment of Chromiferous Iron. — Harry H. Camp-
bell, Steelton, Pa.
795,218. Furnace for Treating Sheet Iron and Steel. — Harry
H. Goodsell, Leechburg, Pa.
795,258. Gas-Producer Apparatus. — Carleton Ellis, New York,,
N. Y., assignor to Eldred Process Company, New York, N. Y., a corpora-
tion of New York.
795,275. Process of Manufacturing Portland Cement from
Slag. — Carl von Forell, Hamburg, Germany, assignor to Henry Ed-
munds, London, England.
795,517. Process of Producing Tungsten Steels. — Edward
D. Kendall, Brooklyn, N. Y., assignor of one half to Edward N. Dickerson,
Stovall, N. C, and one fourth to Emmet R. Olcott, New York, N. Y.
795,643. Mold for Casting Rolls. — Frank M. Newingham,.
Apollo, Pa.
795,835. Gas-Producer. — William B. Hughes, Wissahickon, Pa.
795,842. Tuyere-Bushing. — Frank Klepetko, New York, N. Y.
795,907. Gas-Producer. — Henrich Gerdes, Berlin, Germany^
assignor to American Suction Gas Producer Company, Lansing, Mich.
795,918. Gas-Producer. — Ernst Korting, Pegli, Italy.
798,242. Wire-Annealing Furnace. — John F. Warwick, Chi-
cago, 111.
798,258. Metallurgical Furnace. — George H. Benjamin, New
York, N. Y.
798,500. Gas-Producer. — Carleton Ellis, New York, N. Y.,
assignor to Eldred Process Company, New York, N. Y., a corporation of
New York.
GREAT BRITAIN
12,816 of 1904. Hardening Steel. — S. N. Brayshaw, Manchester,
A bath for hardening steel, consisting of chlorides of potassium and so-
dium, with a small amount of ferro-cyanide of potash.
14,757 of 1904. Purification of Blast-Furnace Gas. — B. H.
Thwaite, London. Improvements in the inventor's process for purifying
blast-furnace gases and making them available for use in gas-engines.
384
ROBERT FORRESTER MUSHET
SEE PAGE 443
The Iron and Steel Magazine
•* Je vetix au mond publier
d'une plume de fer sur un papier d'acier."
Vol. X November, 1905 No. 5
OVERHEATED STEEL*
By ARTHUR WINDSOR RICHARDS and JOHN EDWARD STEAD
Introduction
TX view of the fact that there still exists much confusion in
^ the minds of many metallurgists as to the definition of the
term overheating as applied to steel, and as to whether or not
reheating overheated vSteel can invariably be relied upon to
restore good properties to such brittle material, we venture
to present a review of the opinions of those who have specially
studied the question, and to supplement our previous work
on this subject by further experiments in the hope that some
of the confusion may be removed.
The pioneer work of Tschernoff, afterwards confirmed by
Brinell and others, shows clearly the value of proper heat treat-
ment in restoring forged steel, which had been made coarsely
ciystalline by heating at a high temperature, into material of
finely crystalline character and good physical quality as was
determined by the ordinary system of testing.
Professor Howe, in his " Metallurgy of Steel," states that
" steel which has been exposed to a very high temperature is
known as ' burnt.' It is cold-short and brittle, can be forged
and welded only with care and has a low tensile strength.
Its fracture is coarse and even flaky, crystalline, with brilliant
facets. Steel known as ' overheated ' has a coarse structure,
which may Vje removed more or less completely by reheating
or careful forging. Excessively long or strong overheating pro-
* " Iron and Steel Institute," September, 1905.
386 The Iron and Steel Magazine
duces the structure known as ' burnt,' and the coarseness and
brittleness due to burning are removed with greater difficulty
and much less completely than those due to overheating, yet
in the same manner and by the same expedients."
The same authority, in his work on " Iron, Steel and other
Alloys," states: "As a palliative for burning, mechanical
refining by rolling, etc., is much more effective than heat refining,
as we should naturally expect. For while heat refining should
be powerless to close up even the most minute cracks, the com-
pression and kneading which accompany mechanical refining
should have a powerful effect in closing cracks even of consider-
able size, especially if their sides have not become coated with
iron oxide."
William Metcalf, in his book, " Steel: A Manual for Steel
Users," says: " A ' fiery ' fracture indicates too much heat. It
may be found in the best steel and in the poorest; it may be
corrected by simply heating to a proper temperature. It shows
that some one needs to be reprimanded for careless work." He
further states: "Actual burning reveals itself in rough tears
and cracks at the surface and comers of the piece. Such a
piece should go to the scrap-heap. Overheated steel of coarse,
fiery grain has been injured, and not necessarily destroyed.
Such a piece may be restored to any fineness of grain by heating
to the right temperature — medium orange for the best grain —
keeping it at that heat for, say, one minute for a little piece, and
five to ten or fifteen minutes for a large piece. The heat should
penetrate the whole mass, and it should not be aUowed to run
above the given color in any part, not even for a moment. It
should then be allowed to cool in a dry place, without disturb-
ance. The grain will now be fine and uniform, and the steel may
be worked in the ordinary way. This simple operation is all
that is necessary to restore to a fine grain any piece of steel that
has been overheated, provided that the piece has not been
actually burned nor ruptured. ' ' In the glossary at the end of Mr.
Metcalf 's book is given the following definition of overheating:
" Overheated. — Steel that has been heated too hot, and not
quite burned; its fiery fracture exposes it. The grain of over-
heated steel may be restored, but restored steel is never as
reliable as steel that has not been overheated. Overheating is
a disintegrating operation."
OvcrJicatcd Steel
387
A committee of the Iron and Steel Institute on the nomen-
clature of metallography define " overheated " (Ger. Ueberhitzt,
Fr. Surchauffe) as " Applied to steel that has been heated to
excess and not quite burned."
In Prof. Alfred Stansfield's paper on " The Burning and
Overheating of Steel," * overheating is defined as " reheating
to below the point of incipient fusion," and that " steel that has
been merely overheated can be completely restored by heating
just above its highest recalescence point and allowing to cool."
Professor Heyn, Mr. C. H. Ridsdale and one of the present
authors have shown that ov^erheated soft steel can be and is
completely restored by reheating to a proper temperature above
the point AC3.
In the sixth report of the Alloys Research Committee, by
Sir W. Roberts-Austen and Professor Gowland, there is evidence
that overheated steel can be completely restored to good quality
by reheating.
Professor Arnold, f in discussing the question of restoring
the good properties by reheating brittle steel, claims that it
was invented in Sheffield in 1820, but that those who were best
capable of judging did not call it restoring but " faking.''
Mr. Rogers' J most valuable researches on " The Fatigue of
Steel," a work which must be considered of the very highest
value to engineers and metallurgists, indicates that steel with
0.27 per cent carbon, after overheating at 1215° C. for three
hours and reheating to 900° C. for ten minutes and cooling in
air, compared with the normal steel after annealing it at 655° C,
for half an hour, was restored to a higher fatigue-resisting
character. The results given are as follows :
Reversals to pro-
duce fracture. .
Ratio
Normal Annealed at
6550° C. 17.78 Tons
Fiber Stress
1,493,600
I
Overheated at
1215° C.
16 Tons Stress
636,950
0.42
Restored. 17.78 Tons
Stress
2,692,700
1.80
* " Journal of the Iron and Steel Institute," 1903, No. 2, p. 433.
t Sixth Alloys Research Report. Proceedings of the Institution of
Mechanical Engineers.
X F. Rogers, " Heat Treatment and Fatigue of Steel." " Journal of
the Iron and Steel Institute," 1905, No. i, p. 486.
388 The Iron and Steel Magazine
The bars as rolled, however, before annealing for a short
time at 650° C, stood more fatigue than the restored steel,
inasmuch as it took 2,630,100 reversals at 19 tons stress to pro-
duce fracture, while the restored steel broke down with 204,350
reversals at 19 tons stress.
The experience of all steel founders has proved beyond
doubt that their castings, when they leave the molds, are
very coarsely crystalline and more or less brittle, and that
reheating to between 800° and 900° C. destroys the coarse
crystallization and removes the brittleness. The physical
character of overheated forged steel is identical with that of
steel castings, excepting that the latter are liable to contain
intercrystalline deposits. One w^ould naturally expect that
the same heat treatment would be followed in each class of
material, with the same improvement.
It is well known that in the production of armor plates the
steel, during its prolonged sojourn in the cementation furnace,
becomes exceedingly coarsely crystalline and brittle, and in
such condition may be regarded as overheated, and that by
proper reheating the coarse brittle character is completely
removed, and a steel of exceptional toughness produced.
Capt. H. Riall Sankey and Mr. J. Kent Smith in a recent
paper * show that after overheating their chrome- vanadium
steel at 1200° C. for twelve hours and reheating to 950° for
half an hour the mechanical properties, according to all the
tests applied, compared with those of the raw steel, were much
improved.
In our previous paper f it was shown that, provided the
heating was not carried to the point of disintegration, over-
heated steel could be completely restored to excellent quality,
and made even superior to what it was in the forged condition.
In traversing the expressed conclusions of the authorities
quoted it seems evident that both Professor Howe and Mr.
Metcalf, although they are emphatic in their statements that
" heat refining " removes coarse and produces fine crystalliza-
tion, are not so strongly convinced that the steel will be com-
* " Heat Treatment Experiments with Chrome- Vanadium Steel."
Proceedings of the Institution of Mechanical Engineers, 1904, p. 13 19.
t " Restoration of Dangerously Crystalline Steel by Heat Treat-
ment." " Journal of the Iron and Steel Institute," 1903, No. 2, p. 119.
Overheated Steel 389
pletely restored; indeed, Metcalf states that " restored steel is
never so reliable as steel which has not been overheated." The
remark, however, that " overheating is a disintegrating opera-
tion " clearly shows that what is meant here by overheating is
incipient disintegration.
Professor Stansfield's simple definition of overheating is
practically that which we hold, qualified by the obvious inference
that until the point of incipient fusion is reached there cannot be
any disintegration.
High-carbon crucible steels, such as were available at the
time when Mr. Charles Wardlow made his experiments in
Sheffield in 1820, are very easily incipiently disintegrated by
heat, as is proved by the fall in the specific gravity produced
by high heating. Reheating will not perfectly restore it unless
it is forged to a smaller size, and not even by forging if the steel
has suffered severe disintegration.
The confusion existing appears to have arisen by con-
founding " burnt " with '' overheated " steel, and it certainly
seems to us that this will disappear if the following definitions
of overheating and burning are accepted, viz. :
Overheating is heating at any point below that which pro-
duces incipient disintegration and results in the formation of
large crystals.
Burning is heating at or above the point at which such
disintegration occurs; burnt steel is nearly always coarsely
cr\"stalline.
If these definitions are accepted, the following facts should
be remembered ;
(i) That all overheated steel is more or less coarsely crystal-
line.
(2) That different steels, apparently of the same composition,
vary in their susceptibility to disintegration. At a given high
temperature one may be simply overheated whilst another may
be burnt and partially disintegrated.
(3) That burnt steel cannot be completel}^ restored by re-
heating; it can be greatly improved, but is never equal to re-
heated steel which has not suffered partial disintegration.
As an example of such burnt steel before and after heat
treatment photomicrographs. Figs. 4 and 5, are given, in which
390
The Iron and Steel Magazine
it is shown that some of the broad ferrite sheets or bands remain
after reheating, and these are divided centraUy by some sub-
stance quite foreign to the steel proper. On attempting to bend
this steel the ferrite bands broke up along the line or plane of the
foreign substance, and this, of course, led to sudden fracture
through the whole mass. Such burnt steel is only lit for the
scrap-heap or remelting; it is worthless until it has again passed
through the fluid state.
General Description of Work Done
In this research, instead of working upon rails we have
mainly confined our work to the treatment of i-inch square
rolled bars. The tests applied were as follov/s :
(i) Testing for m^aximum tenacity and elongation in a 50-
ton Buckton machine having a hydraulic cylinder driven from
an accumulator and belt-driven pump coupled to a motor
running at a constant speed.
The ^deld point was observed by means of a pair of callipers
adjusted to suitable marks on the test-piece. The diameter of
Test- Piece
Steel Crip
Steel Crip
the bars, turned parallel, was 0.75
inch, and the portion subjected to
tension 4 inches in length.
(2) Ordinary bending tests.
The pieces tested were cut from
the bars and were highly polished,
being of the following dimensions :
Length, 100 mm.; breadth, 6
mm. ; thickness, 3 mm.
The measurements recorded
in the table of tests represent the
radius of the convex side of the
bend when fracture occurred.
It will be understood that the
greater the radius the less ductile
the steel, and vice versa.
(3) Alternate bending of
strips, having a section 6 mm.
X 3 mm., through an angle of
40°. The number of bendings backward and forward to pro-
duce fracture was recorded in each'case. (See Fig. i.)
Fig. I. Bending to Fracture
I
Overheated Steel
391
(4) Rotary bending through a constant slight angle, of
round pieces of the same dimensions as those used for fiber stress
testing described in (5).
(5) Testing by reversals of stress, by the Wohler method,
on a machine built by Messrs. Richardson, Westgarth & Com-
pany to designs supplied by us, it being an improved form of that
described in our previous paper. (See Fig. 2.)
The machine consists of a strong steel spindle on which is
mounted three pulley wheels, giving speeds of 800, 1,200 and
End of Shorr Bar rests in a hardened steel groove at end of
lever pivoted on a Knife Edge at tne point A From ttie opposite
end of tf:e Lever hangs suspended an adjustable weigtit
Test Piece
Fig. 2. Richards and Stead's Machine for Testing Steel by
Reversal Stress and Strains
2,400 revolutions per minute when driven by a suitable motor.
The spindle moves in massive bearings, and each end terminates
in a modified form of self-centering chuck in which the test-
pieces are securely fastened.
To one end of the machine is attached a short steel bar 10
inches long, also with suitable chuck, which is connected to the
spindle by the test piece, one end of the bar running free, and
having a groove in which is loosely fitted one end of a hardened
steel lever carrying the load used in testing. The other end of
392
The Iron and Steel Magazine
the machine is similarly fitted, the bar being 42 inches long, and
Bends-Series F
Fl
Normal
rr\
F2
Overheated
F3
Reheated
rK\
HJ
F4
Annealed
Bends-Series H
/R\
u
H r.
Normal
H 2
Overheated
H 4
Annealed
Bends-Series C
Peheated
C4
Annealed
the load is applied by means of a spring balance, following the
plan of Professor Ewing and Mr. Humfrey. A suitable counter
Fig. 4.
Overheated Steel 393
for recording the number of revolutions is also attached. (See
Fig. 3-)
The test pieces were uniform in size, being 4 inches long, and
turned to a diameter of one-half inch, the center of each being
further reduced by a semicircular groove to a diameter of
exactly i cm. (See Fig. 3.)
Analysis of Steels
Series F Series G Series H
Per cent Per cent Per cent
Combined carbon 0.06 0.48 0.44
Manganese 0.20 0.82 0.80
Silicon 0.02 0.045 0.02
Sulphur 0.05 0.06 0.064
Phosphorus 0.032 0.087 0.054
ml^ , ,
5urnt Rail Steel. Magnified Fig. 5. Burnt Rail Steel. Same as Fig. 4,
50 diameters. after reheating to and cooling from
850° C. Magnified 50 diameters.
Description of Treatment and of the Terms Used
(i) Normal Steel. — Bars tested in the condition as received
from the rolls.
(2) Overheating. — The bars were packed in quartz sand in a
closed firebrick tube and heated' at a temperature of about
1300° C. for from two to three hours. The tube with con-
tents was allowed to cool in air before opening.
394
The Iron and Steel Magazine
(3) Reheating. — The overheated steels were packed in a
closed firebrick tube and heated in a muffle furnace until a
temperature of 880° C. was reached, with the exception of
Series F, which were heated to 950° C. The tubes with their
contents, as soon as the desired temperature was reached, were
removed and allowed to cool in air before opening.
(4) Annealing. — The rolled bars were slowly heated in a
muffle furnace to a temperature of from 850° to 880° C, with the
exception of Series F, which was heated to 950° C, and were then
withdrawn and allowed to cool in air.
(5) Sorhitic Treatment. — Portions of the rolled bars of
Series H were heated to 900° C, quenched in cold water and
afterwards reheated to 330° C. for twenty minutes and allowed to
cool in air.*
Tensile Tests
Ratio of
Yield
Ultimate
Yield
Elonga-
Con-
Point.
Stress.
Point to
tion.
traction
us
Tons per
Tons per
Ultimate
Per Cent
of Area.
_JJ
Square
Square
Stress.
in 4
Per
Inch
Inch
Per Cent
Inches
Cent
F I
Carbon,
0.06 per
cent normal . . .
16.05
24.40
65.70
35-00
66.30
F 2
> >
) )
overheated
14-75
23.80
62.00
33-00
60.80
F 7,
> >
)>
reheated .
15-85
24.20
65-50
40.00
67.10
F 4
> >
> J
annealed .
15-65
23.60
66.30
37.00
70.90
G I
Carbon,
0.48 per
cent normal . . .
27.20
45-80
59-70
19.00
34-50
Gr 4
>>
)>
overheated
27.40
46.90
58-50
10.00
26.50
G ^
» >
» >
reheated .
27.00
45-80
59-00
22.00
38.60
G 2
) )
>>
annealed .
22.00
48.30
55-30
21.00
36-50
H I
Carbon,
0.44 per
cent normal . . .
21.70
40.20
53-80
23.00
48.80
H 2
> 1
) )
overheated
20.60
41.30
52.40
12.00
20.60
H3
> >
»>
reheated .
22.00
40.20
54-70
25.00
51.10
H4
,,
>>
annealed .
21.30
39-90
53-40
24.00
50.00
H5
> )
> >
sorbitic . .
43-40
55-To
78-70
25.00
56-80
Ordinary Bending Tests
Section of bars 6 mm. by 3 mm.
F G
Normal radius of bend 6.50 7.25
Overheated ,, ,, 7.00 10.80
H
8.00 nnn.
T2.20
* Although this treatment is very liable to produce water cracks,
none were produced in our experiments. Further triads are in progress to
ascertain the best possible method of producing sorbitic steel.
Ovcrlicatcd Steel
395
F G H
Reheated radius of head 6.00 6.65 6.40 mm.
Annealed ,, ,, 6.00 6.80 7.00 ,,
Sorbitic ,, ,, . . . . 6.00 ,,
(See Fig. 3.)
At 6 mm. radius the bends were quite close on the concave side.
Alternate Bending of Strips through an Angle of 40 Degrees
F I
F 2
F 3
F 4
G I
G 2
G 3
G 4
H I
H 2
H3
H4
H 5
Average Bends to
Comparison.
Produce Fracture
Normal = i
Carbon,
0.06 per cent normal
48
1. 0000
>»
,, overheated. .
36
0.7500
>>
,, reheated . . .
53
I.I04I
) >
,, annealed . . .
52
10833
Carbon,
0.48 per cent normal
115
1. 0000
1 *
,, overheated. .
51
0.4434
ft
,, reheated . . .
116. 3
I.OII3
1 >
,, annealed . . .
II5-3
1.0026
Carbon,
0.44 per cent normal
117
1. 0000
,, overheated. .
72
0.6125
, >
,, reheated . . .
132
I.1271
,, annealed . . .
129
1. 1000
M
,, sorbitic . . . .
263
2.2444
Resistance to Alternating Fiber Stresses
Revolutions per minute, 1,200
at
Total Revolu-
tions to Produce
Comparison.
Fiber Stress
0
C/3
Fracture
Normal = i
F 1
Carbon,
0.06 per cent normal . . .
13,560
1. 0000
25 tons
F 2
> ,
,, overheated
9,760
0.7197
25 ,.
F 3
>>
„ reheated .
14,500
1.0693
25 ,.
F4
> >
,, annealed .
14,200
1. 0471
25 .»
G I
Carbon,
0.48 per cent normal . . .
83,550
1. 0000
28 tons
G 2
> ,
,, overheated
38,660
0.4627
28 „
G 3
,, reheated .
111,500
1.3226
28 „
G4
> t
,, annealed .
103,130
1.2344
28 „
H I
Carbon,
0.44 per cent normal . . .
1,432,500
1. 0000
25 tons
H2
1 >
,, overheated
844,950
0.5891
25 M
H3
If
,, reheated .
2,080,440
1.4528
25 „
H4
>i
,, annealed .
1,971,000
1-3773
25 M
H5
>f
,, sorbitic . .
3,517,200
2.4570
25 M
396 The Iron and Steel Magazine
Rotary Bending through an Angle of 6 Degrees
Revolutions per minute, 1,200
U)
Total Revolutions
to Produce
Conipari.son.
V
in
Fracture
Normal = I
F I
Carbon,
0.06 per cent normal
3.500
1. 0000
F 2
,,
,, overheated. .
2,590
0.7400
F 3
, ,
,, reheated . . .
3,730
1-0655
F 4
I >
,, annealed . . .
2,680
1. 0510
G I
Carbon,
0.48 per cent normal
3,940
1. 0000
G 2
,,
,, overheated. .
2,040
0.5177
G 3
, ,
,, reheated . . .
5.300
1-3451
G 4
"
,, annealed . . .
5,120
1.3000
H I
Carbon,
0.44 per cent normal
3,660
1. 0000
H 2
I )
,, overheated. .
2,207
0.6306
H3
,, reheated . . .
5.390
1-4731
H4
, ,
,, annealed . . .
5.215
1-4255
H5
) >
,, sorbitic ....
9,280
2-5361
In reviewing the results of our experiments, what appears to
be most remarkable is the fact that the indications obtained by
all systems of testing prove that overheating reduces the power
of the steel to resist fatigue ; that reheating such steel more than
restores the original good qualities of the rolled bars, and that
when the steel has the carbon in the sorbitic condition its power
of endurance is more than doubled.
The results amply confirm those published in our previous
paper, " Heat Treatment of a Broken Axle " (Series i).
By the kind permission of Mr. James Holden of the Great
Eastern Railway Company we are able to describe experiments
made to determine the effect of heat treatment on a wagon
axle which broke at a flaw after being in use for twenty years.
The microstructure indicates that it had been initially slightly
overheated. The analysis rnade by Mr. J. H. D. Jenkins,
chemist to the Great Eastern Company, of a portion of steel
cut from the axle was as follows :
Carbon 0.340 per cent
Manganese 0.837 ,,
Silicon 0.053 >»
Sulphur 0.047 >)
Phosphorus 0.085 >>
Overheated Steel
397
Portions were cut from the axle under the wheel seating,
where the stresses during the life may be assumed to have been
inappreciable, and from near to the central axis where the fatigue
stresses must have been at a maximum and near to where the
fracture occurred. One portion of each was reheated in a
muflle furnace to 820° C. and was cooled in air; these, with other
untreated portions, were tested in a tensile testing machine at
the Stratford works of the Great Eastern Railway Company
with the following results, viz. :
Tensile Test
Ultimate stress ...
Elongation per cent
in
inches
Reduction of area
per cent
Central Axis
Untreated
37.83 tons
27.08 per cent
47-37 ».
Reheated at
820° C.
From under
Seating.
Untreated
39.28 tons 38.31 tons
29.17 per cent 27.08 per cent
49.68 „ I 45-15 »
Wheel
Reheated at
820° C.
39.71 tons
28.12 per cent
50-32 „
Small pieces from each portion were treated as described
below, and were subjected to continuous reversals of fiber
stresses in the machine described in our previous paper.
Resistance to Reversals of Stresses
SAMPLES FROM CENTRAL AXIS
Alternation of
19 Tons Fiber Stress
Norma] 31 1 ,400
Reheated to 820° C. and cooled in air 1,506,000
Reheated to 850° C. and cooled in air 2,526,000
Reheated to 870° C, cooled in air, and re-
heated again to 850° C. and cooled in air. . 13,532,000
SAMPLES FROM UNDER WHEEL SEATING
Normal H^^ ^^^'°°°
( (b) 208,000
Reheated to 820° C. and cooled in air 1,052,000
Reheated to 850° C. and cooled in air 13,630,000
Reheated to 870° C, cooled in air, and re- f 11,630,00c at 19 tons.
heated again to 850° C. and cooled in air I 2,833,000 at 21 tons.
In each case the pieces of metal reheated did not weigh more
than four ounces, the heating was rapidly effected and as soon
398 The Iron and Steel Magazine
as the desired temperature v/as reached they were removed
from the furnace. It is probable that a longer heating at
somewhat lower temperatures would have given equally good
results.
As the pieces heated were small, and the cooling necessarily
comparatively rapid, the treatment maybe regarded as approach-
ing that of oil tempering on large masses.
In reviewing the known data received from Mr. Holden
about this axle, together with the information obtained by our
own testing, the following observations should be noted, viz. :
(i) The axle stood twenty years before it finally broke.
Fig. 6. Fractured Axle. Magnified i^ diameters.
Supplied by Mr. J. H. D. Jenkins.
During this period it had traveled probably 300,000 miles, and
had been subjected to 200,000,000 reversals of stress.
(2) The flaw which initiated fracture appears to have been
a deep stamp mark, clearly seen in the photograph supplied
by Mr. Jenkins (Fig. 6).
(3) The weakness produced initially by this deep impression,
in our opinion, eventually led to a fatigue fracture, which
traveled from the upper to the convex termination of what
corresponds with the light part shown in the photograph.
Sudden or granular fracture followed through the remaining
part of the mass.
Overheated Steel 399
(4) The interence is that had there been no such excessively
deep stamp impression the axle could not have failed.*
(5) According to the fiber stress testing, the normal steel of
rather coarse crystalline structure is relatively weak, but, for
all that, it is good material, and calculated to be quite strong
enough to stand all the normal working stresses.
(6) Proper heat treatment very greatly increases the fatigue-
resisting properties of the steel.
(7) That ordinary tensile tests which do not show the
yield points fail to give a hint as to the relative value of steels
in their power to resist fatigue.
The photomicrographs. Figs. 8 and 9, represent the structure
of the steel before and after heating to 870° and 850° C.
Hypothetical Conclusions
It is impossible to refrain from endeavoring to explain
how it is that overheated steel should be so weak and non-
resistant to the continued application of reversals of stresses
and strains as compared with the same material in the normal
and restored conditions. In doing this the following facts must
be taken as bases, viz.:
(i) That in the normal and restored material the crystalline
structure is in each case fine, whereas it is coarse in the over-
heated steels.
(2) That in Series G, H and I the crystal grains of large
dimensions in the overheated steels were surrounded by more
or less complete envelopes of massive free ferrite, and that in
Series I the grains enclosed by these envelopes were cut up by
plates or sheets of the same substance.
(3) That on bending the polished and etched specimen of
the overheated steel backward and forward it was easily observed
under the microscope that the extension or distortion during
the bending was mainly confined to the massive ferrite envelopes
and sheets. At first they appeared to sink below the surface,
and eventually developed into actual fissures or incipient cracks ;
* It may here be mentioned that, in the experience of one of the
authors, this weakness and eventual fracture in axles is not by any
means uncommon as the result of a deep stamp indentation. Such in-
dentations are the exact equivalents of flaws, but instead of being acci-
dental are deliberately x^roduced.
400
The Iron and Steel Magazine
an observation confirmed by Mr. Jenkins in the case of the
broken axle.
(4) The researches of Professor Ewing, Messrs. Humfrey,
A. E. Seat on * and A. Jude have proved that fracture of steel,
when under the influence of long-continued fatigue or under
sudden shock, is initiated in the free ferrite, and generally travels
continuously as far as possible along a track in which there is the
greatest amount of that constituent, in fact along the line of
least resistance.
Fig. 7. Portion of Broken Axle; central part of one of the large grains.
Photo supplied by Mr. Jenkins. Magnified 150 diameters.
(5) That in pure or nearly pure commercial carbon steels
the ferrite has a yield point of probably not more than 8 to 1 2
tons per square inch, whereas the other constituent, pearlite,
docs not sensibly yield until a tension of about twice these
amounts is reached, excepting when its structure is composed
of broad sheets of ferrite and carbide of iron, for in such case the
* " Impact Tests on Wrought Steels of Commerce." Proceedings of
the Institution of Mechanical Engineers, 1904, p. 1135.
Overheated Steel
401
ferrite behaves as if it were massive, and the yield point is
necessarily nuioh lower.
(6) The weakness of a steel under stress in any given
direction depends on the manner in which the ferrite and pearl ite
are arranged with relation to the stresses applied. For in-
stance, in large forgings it not infrequently happens that the
ferrite and pearlite are arranged in more or less continuous bands
or sheets. If a polished and etched section is bent slightly at
right angles to the plane of these, and is then examined, it will
be found that the ferrite bands are depressed below the surface,
whilst the lines of pearlite remain in relief. On attempting to
Fig. 8. Broken Axle after Heat Treatment;
section cut from a fatigue test piece.
Magnified 50 diameters.
Fig. 9. Broken Axle in Normal State; sec-
tion cut from a fatigue test piece.
Magnified 50 diameters.
bend to any great extent, the piece will break at a very poor
angle. If, however, bending is effected with the fiber, the steel
will bend quite close without fracturing. If such a steel is
heated to a sufficient temperature so as to cause the carbon to be
regularly diffused through what was originally massive ferrite,
and it is then quenched in water and reheated to 650° C, it will
bend close equally in each direction, provided that there is a
complete aVjsence of strings or plates of sulphide or silicate of
manganese.
402 The Iron and Steel Magazine
To make the point clearer by a hypothetical case, let us
assume that a bar of pearlite is divided midway between its
ends by a plate of ferrite i-io inch in thickness, and is per-
fectty united with the pearlite on each side of it. If this bar
were to be submitted to tension, the ferrite would begin to
extend and sink below the surface, and when it had extended as
far as its nature would admit, it would break before the elastic
limit of the pearlite was reached. The extension of the Avhole
bar would be very slight, but that of the ferrite plate would be
probably 20 per cent or more. The tenacity would be that of
the ferrite. In such a case the ferrite would be the exact equiv-
alent to the weakest link in a chain.
With these facts before us, it is not difficult to explain the
weakness of coarsely crystalline structural steels. The com-
paratively massive and continuous sheets of ferrite approach
more or less nearly to the actual and supposititious cases referred
to of ferrite between pearlite. When under stresses which are
below the yield point as determined by the testing machine, but
which cannot be below that of the ferrite, they are concentrated
on these broad bands, movements backward and forv/ard, with
each reversal of stress, occur between the particles of the soft
ferrite, till eventually incipient fracture is started, and, once
begun, it rapidly travels through ferrite and pearlite alike until a
complete separation of the mass is effected. One broad plate of
ferrite with planes at right angles to the direction of tension and
comipression will be sufficient to initiate fracture, and after this
has actually broken there will be a ver}^ w^eak place where the
stresses will be m^ainly concentrated, and the crack will grow
to complete fracture.
In this way is explained why it is that the steel immediately
adjoining a fatigue fracture is usually fotmd to be practically
as good as it was when first put in. The stresses soon find
out the weak spot, and make it weaker and weaker by being
continually concentrated there, whilst the adjoining steel is often
practically not affected or weakened.*
In the case of overheated steels containing above i per cent
carbon, the brittle envelopes of carbide of iron which surround
the grains or which pass through their substance are the places
along which fracture most readily travels.
* There are, as we have proved, exceptions to this rule.
Overheated Steel 403
Overheated steels consisting entirely of pearlite are very
brittle, and probably fracture is initiated in the plates of carbide.
Ingot iron may be made coarse grained and brittle by long-
continued heating at temperatures below Ac3-87o° C, if the
carbon is very low, and if it exceed o.io, by heating at a very
high degree of heat. In either case the steel is weak, and frac-
ture is initiated between the cleavage planes of the ferrite when
subjected to stresses and strains.
On the other hand, when the crystals are fine and the ferrite
exists in minute crystals and are heterogeneously oriented and
are intimately distributed with the pearlite throughout the
mass, the strength of the two constituents is averaged, the
lines of stress pass over a multitude of each of them, and
the pearlite being the stronger, supports the ferrite and pre-
vents it breaking down.
If these conclusions are right, it seems obvious that it would
be an ideal condition if free ferrite were absent in carburized
steels which have to be subjected to severe vibratory stresses.
Ordinary structural steels in this condition can be obtained by
heating to a suitable temperature, quenching and reheating at
a lower temperature. The example described above as sorbitic
steel was prepared in that way. It was devoid of free ferrite,
and it will be seen that it has double the resisting power of the
normal forged bars.
Before concluding we wish to emphasize the fact that we
do not maintain that steel initially bad, brittle and dangerous,
owing to irregularity in the distribution of the elements, or
from other causes which have not yet been explained, can be
made good by any kind of heat treatment. What we believe
has been proved conclusively is that good steels which have
been heated to any point short of incipient disintegration, and
made excessively brittle by such treatment, can be completely
restored to perfectly sound and reliable material. Also, that
it is safest to heat to a temperature about 50° above the critical
points to insure the complete change of every portion of the
steel, excepting in the case of the purest and most homogeneous
steels, when the temperature of the upper critical points need
not be greatly exceeded.
In conclusion, we gratefully acknowledge the assistance of
Messrs. Bolckow, Vaughan & Company, Ltd., who have en-
404 ' The Iron and Steel Magazine
couraged our research, of Mr. James Holden of the Great Eastern
Railwa}^ Company, and his chemist, Mr. Jenkins, for their con-
tribution and permission to pubHsh the experiments made on a
waggon axle.
We must also acknowledge the valuable services of Mr.
R. C. V. Whitfield, who is responsible for practically all the
experiments on heat treatment and the mechanical testing.
NEW GIN PROCESS FOR THE ELECTRICAL MANUFAC-
TURE OF STEEL*
By GUSTAVE GIN
THE new electric furnace which I have devised for the elec-
trical manufacture of steel allows of the simultaneous and
uninterrupted realization of the following operations :
Fusion, oxidation of impurities, reduction of dissolved oxide
of iron, recarburization or introduction of the constituent ele-
ments of the final steel.
The accompanying design represents, to call it an example,
one form of construction of a furnace accomplishing the condi-
tions indicated.
Fig. I is the plan of the furnace.
Figs. 2 and 3 are transverse sections.
The furnace comprises essentially :
1. A crucible for fusion and refining by oxidation (i).
2. A compartment for reduction and recarburization (2).
3 . A chamber for the observation of the color.
The electrodes of compartment (i) are connected to one of
the terminals from the source of electricity, and the electrodes of
compartments (2) and (3) are connected in parallel with the
other terminal.
The current passes from the electrodes to the metal through
a sheet of slag, which is the principal seat of the heating action
of the current.
The baths of metal communicate by the openings (B) of
sufficiently reduced section, in order that by reason of the Joule
effect the metal may not remain solid.
* American Electrochemical Society, Bethlehem, Pa., meeting,
September, 1905.
A'rzi' Gin Process for the Electrical Manujacturc of Steel 405
The fitting of the compartments is appropriate to their func-
tion: The base and sides of the oxidation compartment (i) are,
in the part permanently occupied by the metal, constructed of a
material basic or acid, according as the metal to be refined is or
is not phosphatic ; but the part in contact with the slag should be
neutral, of chromite of iron, for example.
In compartments (2) and (3) magnesia is employed prefer-
ably for the floor and surfaces not touched by the slag, and
chromite of iron is made use of for the upper part.
To put the furnace into
operation, melted iron or
liquid steel is introduced
through the orifice (A). It
distributes itself through the
three compartments, on the
floors of which scrap iron is
strew^n, the pieces being
placed in connection with (B).
Then arcs are created and
little by little the material is
introduced, which, after fu-
sion, forms the superficial
baths in which the Joule effect
is produced.
The oxidizing bath of
compartment (i) is composed
of some mineral or slag rich
in oxide of iron, in addition
some lime if the metal to be
refined is phosphatic. The baths of compartments (2) and (3)
are neutral and little reducible b}^ carbon.
Aluminates of lime and magnesia, made from a mixture of
bauxite, w4th limestone or dolomite, give favorable results. The
addition of calcium fluoride makes it more fusible and mobile.
The functions of the furnace are characterized as follows:
In the oxidation chamber the zone of most intense heat is, of
course, found close to the contact of the metal and slag, and it is
in this region that the reduction of the oxides is effected at the
expense of the silicon, and of the manganese and carbon of the
metallic bath.
4o6
The Iron and Steel Magazine
U 2
According to the proportion of reducible oxides, the reaction
is more or less rapid, and is evidenced by the turbulence of the
bath, which stirs the metal and aids the oxidation by constantly
changing the surface of contact. The oxidizing action is kept
going and its intensity and rapidity regulated by the careful
introduction of oxide of iron or by stirring.
Because of the high temperature, the elimination of the
carbon is effected almost at the same time as that of the silicon
and manganese.
By means of the method of circulation adopted, the refined
metal in (i) passes next into compartment (2), where the reduc-
tion of the dissolved ferric
oxide and the recarburization
of the metal takes place.
Thus two operations are
accomplished entirely by the
admission of carbon in the
form of coke, of iron or of
casting specially prepared by
the fusion of iron or steel in
the electric furnace in the
presence of a great excess of
carbon. In raising the metal
to a sufficient temperature, a
cast iron is obtained which
absorbs as high as 7 7 per cent
of carbon, which it retains on
cooling, partly as combined carbon and partly as intermingled
graphite. It may be advantageous to introduce this carbureted
cast iron in the liquid state, as when it comes out of the
electric furnace.
The regulation of the recarburization or the incorporation
demanded by the nature of finished steel, is accomplished in
compartment (3), where some test ports are placed which allow
one to estimate the color obtained and indicate additions to be
made in correcting for impurities observed.
The necessary use in the Martin furnace of ferro-manganese
or f erro-silicon , to prevent the oxidation of the metal before the
purification, would be almost useless here. In fact, in the two
last compartments the steel escapes all oxidizing action, for it
New Gin Process jor the Electrical Manufacture of Steel 407
Fp.3
has no contact whatever with the atmosphere, and only comes in
contact with a neutral s\ag. Besides, this slag under the influ-
ence of the high temperature and the presence of carbon furnishes
a slight quantity of aluminum, which diffuses into the bath of
steel and absorbs oxygen in whatever form it might be present.
The charging of the solid material and the withdrawal of the
slag is effected by means of the openings (C) and the stream of
steel by the orifice (D) placed at a given height above the floor.
To tap the metal, the electrodes (3) are lowered and im-
mersed in the metallic bath at the same time that the electrodes
(i) are raised, in order to
maintain the voltage of the
system. During the pouring,
the immersion of the elec-
trodes (3) is increased in such
way that the level of the metal
remains constantly at the
height of the pouring orifice.
After the electrodes touch the
bottom of the crucible, as the
level of the metal can no
longer be maintained, the latter falls until the slag appears in the
flowing jet, which is then stopped.
The operation which has just been explained is of consider-
able importance. In fact, the immersion of the electrodes pre-
vents unequal levels between the compartments (2) and (3).
There can be no mixing of the metals at the different stages of the
refining, and metal can only be run off when refined thoroughly
and to a known color.
The duration of the pouring being nearly constant, the disso-
lution of carbon through the immersion of the electrodes varies
little, and it is possible to keep careful account and guarantee for
the final product a determined color within narrow limits.
After tapping, electrodes (3) are raised again and electrodes
(i) are immersed in the metal; part of the metallic oxide passes
from (i) into (2), while the carbureted steel in (2) penetrates
into (3). Thus the levels remain at their normal height, the
electrodes immersed in (i) occupying simply the place of the
steel which has just been left. If at this moment we tap the crude
iron or incompletely refined steel in the first compartment.
4o8 The Iron and Steel Magazine
raising simultaneously electrodes (i), the metal introduced fills
the space given by removal of the electrodes, but without ability
to penetrate directly into compartment (2), for there is no sen-
sible difference of level. The operations remain there distinct and
independent and can succeed each other in continued rotation as
long as the material retains sensibly the temperature of reduction,
the capacity of the compartments being such that each tapping
only represents a fraction of the metal present, the part remain-
ing playing the part of a heat retainer.
Finally, it can be remarked that the metal in (i) passes to
(2) without carrying the slightest trace of slag, which excludes
all outside impurities.
STEEL AS AN IGNEOUS ROCK *
'T^HE days in which steel was practically regarded as an iron
-*- containing a certain proportion of carbon have long gone by.
We have learned that the other elements always present in steel
exert a decided influence on all the properties of the iron; but
we are still far from understanding the real nature of the purest
steel. While the inexperienced investigator may too readily
imagine that he has struck at the root of the problem. Professor
Arnold, an authority in this field, declared in a discourse delivered
at Johannesburg, before the British Association, on August 31,
that steel was probably the most complex substance extant,
and that he had to confess, after research extending over a quar-
ter of a century, that the more he learned about steel the less he
knew of its ultimate nature.
Considering the vast progress that we have made in testing
materials and the triumphant steel structures which engineers
have erected, this statement may sound unnecessarily pessimis-
tic. The careful designer allows an ample factor of safety, and
the stability and strength of the bold buildings and bridges and
huge ships which have been constructed in the age of steel demon-
strates that the engineer must, to a certain extent, understand
his chief building material. Scientists, Professor Arnold grants,
have perfected the art of steel-making to such a degree that first-
class material will, perhaps, not fail in more than one case out
* " Engineering," September 8, 1905.
StccI as ail Ig}icoiis Rock 409
of ten thousand. But failures do occur, and their mysterious
character proves both that we do not comprehend the real nature
of steel and that our methods of testing are faulty. The first fact
will not be questioned; whenever the ultimate nature is ap-
proached, we have to confess our ignorance. But that our
methods of testing on w^hose perfection so much ingenuity and
money have been spent, should seriously be deficient, will not
universally be conceded.
Homogeneity is the great aim which most metallurgists
have in view. Professor Arnold, as will be seen later, thinks
that safety is to be found in want of symmetry, not in the visible
structure, such as the microscope reveals, but in the molecular
grouping, about which the microscope does not tell us any-
thing. He regards steel as an igneous rock, more or less crystal-
line. To go much further in the definition would, first of all,
require a real satisfactory answer to the question, What is crys-
talline matter? In Professor Arnold's opinion there is a certain
analogy between granite and steel. The granite is built up of
quartz, felspar and mica; the steel of the constituents ferrite,
pearlite and cementite. In an unsaturated steel containing
about 0.5 per cent of carbon, the microscope allows us to distin-
guish patches of iron; the saturated steel, with 0.9 per cent of
carbon, looks more homogeneously speckled with pearlite;
in the supersaturated steel the iron carbide or cementite is seen
to form cell-walls, and the cementation proceeds from outside.
When the rail fracture occurred which caused the St. Neot's
accident some years ago, it was suggested that manganese mono-
sulphide might be responsible, as the steel contained 0.09 per cent
of sulphur. But this sulphur was found to be uniformly dis-
tributed through the rail section in small patches.
Micrographic testing is held in high esteem, and nobody
wishes to derogate its importance. But it cannot always guide
us. When the carbonist theory had been abandoned, great
hopes were placed on the allot ropic theory. The iron of the a
range, up to a temperature of about 740° C, was assumed to be
soft ; in the /? range the iron was alleged to be hard ; in the )-
range, flint hard. The transition ranges seemed to coincide
with the critical temperatures, at which recalescence and change
in magnetic properties are observed. Experimenting on an
ingot of a very pure steel, however. Professor Arnold found that
4IO The Iron and Steel Magazine
the alleged mechanical influence of allotropy did not exist. The
ingot was rolled down, and the test-pieces, quenched at various
temperatures ranging up to 900° C, were etched in nitric acid.
The critical points did not affect the mechanical strength. The
increase in tenacity began 200 degrees below the critical point,
and for the range 500° to 900° C, the tenacity increased pro-
portionally to the quenching temperature. The tests con-
corded so well that the deviations from the curve did not in any
case amount to more than 0.167 ton per square inch, and were
in most instances much smaller.
After this theory had been withdrawn by Osmond in 1901,
polymorphic changes in the iron were suggested. Osmond re-
duced ferrous chloride by hydrogen, and collected crystals of
the reduced iron on porcelain disks inserted in the porcelain tube.
The looked -for rhombic crystals were not found. All the crys-
tals were cubes and octahedrons, internally symmetric, though
interfering with one another as to their external geometry. There
was no trace of rhombic crystallization, and the whole problem
was practically brought back to the state in which Clifton Sorby,
of Sheffield, left it in 1863.
In spite of all chemical, mechanical, thermal and micro-
graphic testing, strange failures were reported from time to time.
One of the most characteristic in Professor Arnold's experience
concerns the following case: The boiler of a warship had been
passed, but later two plates split right across in a line nearly
through, not quite parallel to the seam. The chronology of
these testing operations is the following: Hydraulic pressure
of 228 pounds, 260 pounds and 305 pounds per square inch was
applied on February 5, 8 and 19; on February 20 steam pres-
sure of 60 pounds was tried; on February 21 the boiler burst
under an hydraulic pressure of 270 pounds. On closer ex-
amination it resulted that cracks were also developing between
the bolts. The boiler was very carefully taken to pieces, the
rivets drilled out, and the plates — i inch thick and weighing
3?; tons — were straightened. During this straightening one
of the plates split right across into three pieces, the other into
five pieces. Samples planed off from the fractures proved
most excellent. (See " Engineering," page 164, ante.)
A microscopic comparison was then made by Professor
Arnold between this boiler steel and a sample of the best steel
Steel as an Igneous Rock 411
obtainable. The pearlite fragments in the boiler steel were
more coarse and an,e^ular, the ferrite drawn out into long lines or
ghosts; but there was nothing to account for the fracture, and
the static tests gave no clew. Professor Arnold then designed
a new testing machine on Woehler's lines. The test rod, of
about 4 inches in length, is held by a die which grips a length
of about I J inch; the other 3 inches project, and a reciprocating
plunger takes hold of the free end and deflects it by f inch at
each alternation. The standard machine of this type for bolts
up to 4J inch in diameter applies about 650 alternations per
minute, and a good steel should stand 300 alternations. The
fracture ensues along the die line. Under these alternating
stresses, the fracture starts on the skin, on one or on both sides,
and proceeds to the middle, where a filament is left, which finally
breaks by tension. When both sides give way at the same
moment, the filament will occupy the mid-position; if one skin
gives way before the other, the filament will be found nearer
the other side, which was later attacked.
When mentioning these tests for the first time at the Cam-
bridge meeting of the British Association last year. Professor
Arnold expressed the opinion that the outside and inside parts
of the plate varied in their resisting stresses. This opinion he
now withdraws. The strength really varies all through the
material. This is shown by tests conducted under different con-
ditions, with rates of alternations of 169 or 266 per minute; six-
teen test-pieces from the outside of the plate broke after an
average of 900 alternations (the extremes being 1,292 and 390
alternations), and sixteen pieces from the inside after 839 alter-
nations (the extremes being 230 and 1,177). In both cases the
different test-pieces behaved very differently, therefore, and
there was no reason to give the preference to the outside or the
inside pieces. The best obtainable steel was submitted to simi-
lar tests; complete fracture ensued after about 1,375 (fi'oiii i»33^
to 1,456) alternations at the rate of 168 alternations per minute,
and after 878 (from 860 to 916) alternations at the 266 rate;
this steel was therefore much more uniform. Heat treatment
was tried on the burst boiler steel. The treatment comprised
oil- and water-quenching and annealing at several temperatures,
but the material was not improved. Test -pieces which had
failed after alternations of 230 and 1,240 (the extreme cases)
412 The Iron and Steel Magazine
were then polished just below the fracture line and microscopi-
cally examined; they showed the same appearance, and all these
test -pieces stood the cold bending tests remarkably well.
Somewhat similar experience was gained with an old boiler
whose end plate had cracked right across. In that case also the
ordinary tests could not account for the accident. The new line
of thought which had suggested itself to Professor Arnold was
finally strengthened by another observation. A garden gate had
been held by a bolt of wrought iron, driven through a stone pillar.
The bolt was hammered out of the pillar, and the projecting end
broke off at the fourth blow of the hammer. Microscopic ex-
amination revealed the existence of a crystal which had cleft
parallel to a cubical face; as iron is opaque, the cleavage could
not directly be shown in the ordinary wa}^ on a slide. But the
cleavage forms a series of steps, and the light reflected from
those faces marks the lines of cleavage. If such cleavage can
take place, we can conceive why steel sometimes breaks more
like glass than like a ductile metal. Cases of this kind have
hitherto been explained as results of fatigue; but this is only a
convenient term coined to mask our ignorance.
The tentative conclusions at which Professor Arnold has
arrived in explanation of sudden strange fractures of engine,
boiler and structural steels is, that after the gross crystalliza-
tion, discernible by the highest microscopical powers, has been
completed on cooling, there set in, from a series of centers, mo-
lecular movements tending to the production of perfect mineral
cleavage. This cleavage cannot easily be detected by the micro-
scope, as steel is absolutely opaque to transmitted light. To
avoid the development of cleavage planes the molecular struc-
ture of the steel should be asymmetrical. That crystallization
proceeds from a series of centers can be demonstrated in many
substances, and the decomposition and disintegration will start
from the same centers.
As regards testing, Professor Arnold would not rely too
much on dynamic and alternating stress methods, by which en-
gineers are now guided to a considerable extent. They might
land the engineer out of the frying-pan into the fire, he was afraid,
for they indicated that steel with a high elastic limit was less
liable to fracture under alternating stresses than a metal with a
lower limit. Researches which he had carried out at Sheffield,
Metallography Applied to Foundry Work 413
at any rate, had convinced him that under severe alternating
stresses steel with a high elastic limit was, as a rule, more liable
to rupture than softer low-limit steel. Nor did Professor Arnold
wish to alarm the engineering profession; for these mysterious
fractures are, fortunately, very rare.
Whatever we may think about the best method of testing
and about the relevancy of the cleavage hypothesis, nobody
will doubt that continued conscientious and systematic study
of the steel problems is urgently required. It is satisfactory
to state that the metallurgical department of the University of
Sheffield, which Prof. T. O. Arnold has been able to create,
is at present probably the best equipped in the world, and that
it enjoys the sympathetic support of manufacturers, who enable
their students to join the evening classes, in which they receive
the same instruction as the regular day students. Chemistry,
physics, geology, mineralogy and micrography are compulsory
subjects at Sheffield, and our readers will grant that in the light
of these novel researches none of these subjects could be dis-
pensed with. If Great Britain is to maintain or to regain her
supremacy in armor-plate and ordnance manufacture, — certain
indispensable processes are not of British origin, — such institu-
tions should be provided with ample funds from official and
private sources.
METALLOGRAPHY APPLIED TO FOUNDRY WORK *
PART IV
By ALBERT SAUVEUR
The Photomicrography of Prepared Samples of Cast Iron
TX the preceding installments of this article I have briefly
described the manipulations necessary to prepare samples of
cast iron for microscopical examination, as well as the needed
apparatus. To conclude the technological part of my subject,
I still have to outline the photography of the magnified images
of the samples. The desirability of securing photomicrographs,
and, therefore, permanent records, of the structures under
examination need not be insisted upon. The various steps once
mastered, the operation becomes, moreover, a simple and inex-
pensive one.
* " The Foundry," October, 1905.
414
The Iron and Steel Magazine
Apparatus. — The camera shown in Fig. i as well as in Fig.
6 of the last installment (" The Foundry," page 7, September,
Fig. I
The foundry
1005) gives excellent satisfaction. It is used vertically and if
the microscope be alwavs kept in the same position on the base ot
the camera, all that is needed when it is desired to take a photo-
Meiallography Applied to Foundry Work 415
graph is to lower the bellows so that the light-proof connection
it carries tits into a similar part attached to the tube of the
microscope. Connection between the camera and the micro-
scope is in this way quickly accomplished without disturbing in
the least the specimen or the illuminating parts.
Manipulations. — The taking of a satisfactory photomicro-
graph may be considered as dependent upon the following
factors: (i) The polishing and other preparation of the sample;
(2) the illumination of the sample ; (3) the focusing of the magni-
fied image upon the screen of the camera ; (4) the kind of photo-
graphic plate used; (5) the exposure of the plate; (6) the devel-
opment of the plate and other dark-room manipulations and (7)
the printing from the negative, toning, etc. Unless each of
these factors is given the correct value a satisfactory photomicro-
graph is not to be expected.
The preparation of the samples has been dealt with else-
where and I shall only add here that it is generally advisable to
photograph a sample shortly after it has been etched, the struc-
ture being then generally more sharply defined and brighter. A
slight rubbing of the polished surface over a piece of chamois
leather just before placing it under the microscope also frequently
brightens it and otherwise improves its appearance.
The other factors just indicated as influencing the quality of
the photomicrograph will now be briefly considered in the order
named.
Illumination of the Sample. — Any of the illuminating
apparatus described in the last installment may be used for
taking photomicrographs, but the arc lamp outfit is by far the
most satisfactory. With a weaker source of light, such as a
Welsbach lamp or an electric incandescent lamp, the exposure
requires considerable time, even when using low-power objectives,
w^hile it becomes objectionably long when using high powers.
With high-power objectives, moreover, the amount of light which
reaches the screen, in case of a weak source of light, is relatively
so small as to make a satisfactory focusing of the sample very
diffictilt.
Whatever the source of light used, great care should be
taken to secure as intense and uniform an illumination of the
sample as possible through suitable manipulations of the lamp,
condensing lenses, vertical illuminator, etc. The light-proof
4i6 The Iron and Steel Magazine
attachment should then be inserted and the beUows of the
camera lowered as shown in Fig. i , when the next step will be to
focus the image upon the screen of the camera.
Focusing. — This is probably the most delicate operation
connected with the taking of photomicrographs, and one which
cannot be slighted. Three conditions are essential to success:
(i) The center of the image should occupy the center of the
screen; (2) the image should be evenly lighted; and (3) it should
be accurately focused so as to secure maximum sharpness and
clearness. The image can generally be properly centered by a
slight rotation of the disk of the vertical illuminator as well as by
the rotation of the illuminator itself, while following with the
eye the motion of the image upon the screen.
The absolute necessity of securing a perfectly even illumina-
tion of all parts of the image will be obvious and this is to be
brought about through suitable manipulations of the parts
affecting the path of the beam of light, that is the lamp itself,
the condensing lenses and the vertical illuminator.
The image should be focused as accurately as possible by
gently turning the fine adjustment screw of the microscope, and
with the assistance of a focusing cloth. A focusing glass may be
used with great advantage for this operation and is of special
importance when photographing with high-power objectives. It
consists of a magnifying glass suitably mounted, but an ordinary
eye-piece may be used with nearly as good results. This lens or
eye-piece should be placed on the plain glass circle which occupies
the center of the screen of the camera, and the image carefully
focused while being viewed through this lens. By this means
we magnify the image and are, therefore, able to focus it more
sharply in its finest details. Considerable light, however, is lost,
and the object will generally appear but dimly lighted. The
rule to follow while using a focusing glass is to secure the clearest
possible image while working the fine adjustment tentatively in
both directions, bearing in mind that at its best, the image may
appear blurred and dimly lighted.
When an electric arc lamp is used it is advisable to con-
siderably reduce the light after focusing, by a partial closing of
the iris diaphragm. In doing this we, of course, increase the
exposure time, but as is well known a sharper negative is gener-
ally secured.
Metallography Applied to Foundry Work 417
Kind of Plates. — While any good photographic plate may
be used, I prefer the so-called " process " plate made by Carbutt,
Seeds' " contrast " plates or other similar plates. They are
generally relatively slow but they yield sharper negatives,
having great contrast.
Exposure of the Photographic Plate. — After the image of the
sample has been properly focused the sensitive plate should be
introduced and exposed, with the ordinary precautions, for a
suitable length of time. The time of exposure will vary accord-
ing to (i) the kind of plate used, (2) the illumination and (3)
the nature and color of the sample. The illumination will in
turn depend upon (i) the source of light, (2) the adjustment of
the illuminating parts and (3) the magnifying power and some
other properties of the eye piece and objective.
It is not possible to give positive information regarding the
needed time of exposure. Each student must experiment a
little before being able to judge accurately. As a guide, and
when using the relatively slow plates mentioned above, and a
Welsbach lamp, it may be said that with low-power objectives,
yielding a magnification say of 100 diameters, the exposure will
generally vary between five and twenty minutes, according to
the character of the sample and the care with which the illumina-
tion has been adjusted. Higher powers required considerably
longer time. With a more intense source of light such as that
furnished by a 50 candle-power incandescent lamp, and especially
with that of an electric arc lamp, the time of exposure is, of course,
much shorter, in the latter case varying from a fraction of a
second to five or six seconds under the least favorable conditions.
Development of the Photographic Plate. — While any good
developing formula may be used, my preference is for the old-
fashioned potassium, oxalate-ferrous sulphate developer, using
six parts of a saturated solution of oxalate of potassium to one
part of a saturated solution of protosulphate of iron. To retard
the oxidation of the iron it is advisable to add a few crystals of
tartaric or citric acid to the iron solution. The solutions should
be mixed only as needed. The ordinary intensifiers and re-
strainers may be used in the usual way if the plate shows signs
of having been under or over exposed, but I should advise the
beginner to make another negative in case of decided ill-timed
exposure, as probably the shortest road to success.
4i8
The Iron and Steel Magazine
Printing. — Whether a printing out or a developing paper
should be used for taking prints from the negatives is largely a
matter of individual preference. Personally I prefer the former,
5S
O
Z
z
P
o
o
P5
So
"Z
ei
oi
o
o
u
o
Pi
<
u
<
u
H
S
<
U
Q
W
Z
"Z
w
z
z
Pi
z
S
s
2
^2
o
u
Cfl
o
«■
o
<
3 £
(A! P<
E-i
o
u
t«
s
H^
-c
K
5
o
s
u
w
Q
1
9i
<<
a:
K
H
C3
M
z
H
<;
tn
P^
(d
e
u
M
<
pj
z
o
C-l
fa
o
w
Pi
w
e4
i
z
W
(u
H
o
z
lyj
O
w
|j
0
PJ
o
u
<3
H
>
z
^
H
O
u
V)
s
hH
C/2
tfl
z
D
Z
C/3
r1
z
<
o
Q
<
;^
CJ
w
hJ
hJ
H
Pi
pj
00
H
W
w
P<
H
u
><
^^»B|, .^
X
M&^iw
Pi
W
H"iHv^^
H
St^wi^i^H ^^^^Rj^^
<
j^Sj^^H^^E ^^^BJ|^|
.
Q A
n^HI
k
W3
W
U
«l
PI
1
W
<
Q
W
>
<
W
P-
<
P.
St(||vHH|B^*
*T^Jfc Jjjj^H
f
>
w
pj
P- o
d
Hp^
z
o
<
U
m
O
Ph
o z
z
s
o
w
^
Q W
Fig. 2
but I am well aware that many favor some developing paper.
Drying on ferrotype plates affords a quick means of finishing
silver prints and giving them a satisfactory luster. It is recom-
mended to cut the prints to circles some 2^ inches in diameter
Crystalline and Amorphous States of Metals 419
by means of a margin trimmer and suitable form, as this will give
them a very neat appearance.
Mounting. — The prints should be pasted on special card-
board mounts, such as is shown in Fig. 2, with descriptive matter
and blank spaces to be filled. I also find it very useful to mount
every print on index card, including the same printed and
descriptive matter as that of Fig. 2, stamping upon each sample
the same number as that of the index card, and by keeping the
samples in a suitable cabinet, any desired sample together with
its photomicrograph and accompanying information may at a
moment's notice be taken from the collection. Finally, it is also
advisable to print the same information and to mount a print
on the envelopes or negative preservers in which the plates are
kept. Such system greatly facilitates the work and adds to its
effectiveness.
CRYSTALLINE AND AMORPHOUS STATES OF METALS*
"ITZHEN, a few years ago, Mr. G. T. Beilby published the
first account of his researches upon the hard and soft
states of metals, his work excited more curiosity perhaps than
real interest. It appeared strange that metals should yield to
the gentle pressure of the fingers, and " flow " and transform
their surface layer into a vitreous film. The great part played
by surface tension in liquids did not at once suggest the impor-
tance of these modified surface layers. Yet it should be clear
that if on the surface of solids which have undergone certain
mechanical treatment there is something like the elastic skin
which is apparently stretched over the surface of a liquid, that
fact would strongly influence our tensile strength tests. This
side of the question has, within the last few months, been
touched upon by Mr. Beilby, in a communication which he has
brought before the Royal Society; and the presidential address
which he delivered to the Chemical Section of the British Asso-
ciation at the Johannesburg meeting, on August 29, reviews the
whole problem from a broad theoretical standpoint. Osmond
and others have taken up these researches, which in many
* " Engineering," September 22, 1905.
420 The Iron and Steel Magazine
respects deserve the attention of the engineer. We ha\"e already
drawn our readers' attention to them.
Gold films which are being polished behave exactly like
a liquid under the influence of surface tension. The molecules
tend to heap up in minute mounds or flattened droplets. Some-
times the mounds are so shallow as to become visible under
illumination by an intense oblique beam of light, and such
films mav be not more than 5 or lo micro-millimeters in thick-
ness. They would contain only from 10 to 20 million mole-
cules in their thickness. Moderately thin gold films become
translucent at a temperature of 400° to 500° C, and show by
transmitted light the forms resulting from surface tension. The
astounding malleability and durability of gold is not unlimited.
The finest films of gold and platinum are utilized as electrical
resistances; and when carelessly beaten, gold-foil develops
cracks round its edges. That the original softness can be re-
stored to the gold by beating has long been known to the artifi-
cers. In iron and steel, heat annealing is, as a rule, associated
with the growth of crystalline grains; under over-strain these
grains become deformed by slips occurring along cleavage plam^, ^
as Ewing and Rosenhain have shown. Similar observations have
been made on other metals ; but it is not clear why malleability
and ductility should reach their limit at a point when the crys-
talline grains are, to all appearance, only slightly deformed.
The polished surface film retains no trace of crystalline structure,
and seems to have passed through a liquid condition. It may
be assumed that the conditions which prevail at the outer sur-
face might also occur at the inner surfaces where movement had
taken place, so that every slip of one crystalline lamella over
another would cause a thin film of the metal to pass through
the liquid phase into the non-crystalline condition. Beaten
pure gold-foil reaches its hardest and least plastic condition
only when all trace of crystalline strticture has disppeared.
This state is limited to the surface layers ; for in the interior the
hardened substance produced by the flowing under the hammer
appears to encase and to protect the crystalline units, after they
have become broken down to a certain size. It can be shown
Vjy careful etching that underneath the vitreous surface of the
gold there remains a layer of minute granules, and beneath these
again the distorted fragments of lamellae and grains are met with,
Crystalline aiiJ Amorphous States of Metals 421
embedded in a vitreous and granular matrix. When the metal
is annealed by heating, the crystalline structure is again observed
on etching.
Mr. Beilby's researches prove that the property of passing
from the crystalline to the amorphous phase by mechanical flow,
and from the amorphous to the crystalline phase by heat at a
definite transition temperature, is general, and possessed by all
crystalline solids which do not decompose at, or below, their
transition temperature. He further argues that this change
ranks with the great changes which result in the three generally
distinguished states of aggregation — - the solid, liquid and gase-
ous. All these changes mark alterations in the molecular activ-
ity at certain temperatures and other conditions. On grounds
which require further elucidation he assumes that the amorphous-
crs^stalline change is different from the allotropic changes which
the chemist ascribes to certain elements and compounds. We
know that sulphur occurs both in the rhombic and in the pris-
matic modifications, which differ somewhat in physical and
chemical properties, and which are Vjoth stable within certain
temperature ranges. Both modifications are crystalline, and as
the perfect amorphous state is characterized by the absence of
all crystalline structure, there is a difference between the phe-
nomena. The crystal is a living unit; it can grow, and may be
described as being in a state of dynamical rather than in a static
equilibrium. Below a certain temperature, however, the cr3^stal,
to all appearance, becomes a mere pseudo-morph, with no
power of active growth. But those powers are not extinct;
they are only in abeyance, and ready to be called forth by the
energizing influence of heat. At extremely low temperatures
all chemical affinity seems to become latent.
Hadfield has recently extended to iron and its alloys Dewar's
early experiments on the behavior of metals at the tempera-
tures of liquid oxygen and hydrogen. He conducted observa-
tions at ordinary temperatures and at — 182° C. The tenacity
and hardness of the metal and alloys were invariably enhanced
at the extremeh^ low temperature, and they returned to exactly
their former value when the ordinary temperature was re-
attained. The tensile strength of a pure iron increased from
23 tons at -Hi8° C. to 52 tons at — 182° C; that of gold, from 15
to 22. 4 tons; that of copper, from 19.5 to 26.4 tons. The increase
42 2 The Iron and Steel Magazine
can hardly be ascribed to a closer approximation of the mole-
cules, for the actual expansion coefficient of most metals below
o° C. is extremel}^ small; nor can it be due to permanent changes
of molecular aggregation, for Hadfield obtained a perfectly
smooth and regular cooling curve for iron between +18° and
— 182° C. We must rather suppose that the abstraction of heat
produces a reduction in the repulsive force of molecular vibra-
tion, such that the primary cohesive force can assert itself more
and more.
These researches prove, so far as they go, that the relation
between temperature and tenacity continues unchanged down
to the lowest attainable temperature. Both Dewar and Had-
field tested their metals in the annealed or crystalline condition.
Assisted by his son, Mr. H. N. Beilby, B.Sc, Mr. G. T. Beilby
has attacked this problem from the standpoint of his phase
theory of the hard and soft state of metals. When tenacity is
measured by the tension required to tear asunder a bar or rod, it
is assumed that the tensile stress is uniformly distributed over
the w^hole surface at which rupture ensues ; but this is clearly
not justified. It is impossible to experiment on a single chain
of molecules, and only in a perfectly rigid body could all the
pairs of molecules be pulled apart, as they would be in a single
chain of molecules. If we depart from perfect rigidity, the
molecules under strain will move over each other, and the rup-
ture will become to a certain extent like that of a highly viscous
body (molten glass, e. g.), in which the molecules evade any
direct pull by slipping over each other. In the ductile metals
the crystalline phase is mechanically unstable, while the amor-
phous phase only becomes unstable at a definite temperature.
That the amorphous phase of ductile metals should, on Mr.
Beilby's views, be the hard state, and the crystalline the soft,
seems to contradict the accepted ideas ; for hardness and brittle-
ness are generally associated with the crystalline state.
But Mr. Beilby regards the softest metals as those which
pass most readily into the crystalline conditions, and which are
in their softest stage when in this condition. The softness, in
his opinion, is due to the readiness with which the crystals can
be broken down into the amorphous state. The crystalline state,
in other words, marks the mechanical instability, the amorphous
state the thermal instabilitv- Annealed wires in the crvstalline,
Crystal! i}ic ajid A))iorphous States of Metals 423
or C, state stretch when they are stressed beyond the yield-point ;
hardened wires, practically in the amorphous, or A, state, do not
stretch, but break without extension when their limit of tenacity
is exceeded. The homogeneous C phase of ductile metals has
no true breaking point; it yields and stretches when stressed
beyond the elastic limit, and in doing so passes partly into the
A phase, rupture finally occurring at the breaking point of the
mixed structure. The tenacity of this mixed structure ap-
proaches that of the homogeneous A phase, but does not quite
reach it. A wire which has been hardened simply by stretching
differs from a wire hardened by hammering or drawing, and in
order to obtain the nearest approach to the homogeneous A
phase the C phase is broken down by wire-drawing in Mr. Beilby's
recent experiments.
Wires about .5 mm. in thickness were used. After drawing
the wires through a series of die-plates to four or five times their
original length, all crystalline structure seems to have disap-
peared; yet the wire consists of minute granules of the C phase
embedded in a matrix of the A phase ; the structure is still mixed.
Further drawing at the same temperature alters the structure
only slightly; there appears to exist a certain mechanical equi-
librium between the phases for each temperature. But when
the drawing is continued at lower temperatures, the A phase is
more completely attained. The drawing was sometimes re-
peated until fifteen times the original length had been reached.
Wire-drawing can be overdone, especially at ordinary tempera-
tures — that is to say, the tenacity may decrease. It is evi-
dently advisable to draw wires at the lowest possible tempera-
ture when high tenacity is aimed at, and that tem.perature may
be even below — 182° C. Gold gave the highest tenacity when
stretched to three and a half times its original length.
The experiments so far concern metals of great purity —
that is, gold of 99.97 per cent, silver of 100 per cent, and copper
of a conductivity of more than 100, yet not quite so pure as the
two other metals. In making the tenacity tests, a water load
was applied, so that the speed of loading could be controlled,
and the wire was submerged with its grips, the extension being
measured after taking the specimen out of the liquid air. The
wires broken at ordinary temperature (15° C.) showed no general
stretching. There was a slight extension of .5 or i per cent.
424 The Iron and Steel Magazine
•
entirely due to a sharp reduction of the diameter at the point of
rupture. In liquid air all the wires stretched from ii to 12 per
cent over their whole length between the grips; this was ascer-
ta^ined by careful measurements. The appearance of the frac-
tured ends revealed some interesting points. The ends of the
broken copper wires showed the cupped formation, due evi-
dently to the lower tenacity of the central core, which may be
ascribed to the presence of gas bubbles which the drawing opera-
tion had transformed into long tubes. Some silver wires also
indicated cupped formation; in this case, however, the gas
bubbles did not appear to have been evolved before the moment
of fracture. The gold wires were practically free from spongi-
ness, and the fractures w^ere almost perfectly viscous. The ten-
acity tests yielded the following figures : Gold wires at +15 and
• — 182° C. gave a strength of 15.6 and 22.4 tons per square inch;
silver wires, 25.7 and 34.4 tons; copper wires, 28.4 and 36 tons
at the same temperatures. The figures are not to be regarded
as final, and it looks as if Dewar and Hadfield had also been
working with partly hard-drawn and not with annealed gold.
In the cr^/stalline state the molecules exert their mutual attrac-
tions along directed axes. Further experiments with metals in
the amorphous state may throw light on the qu.estion whether,
and to what extent, the crystalline state depends upon a dynamic
equilibrium between the forces of cohesion and repulsion, or
whether a directed cohesion exists fully developed in the mole-
cules at the absolute zero.
This is a most interesting suggestion. Primitive or blind
cohesion holds undisputed sway at absolute zero temperature,
while with rising temperature the repulsion due to the molecular
vibration becomes more and more strong. The interplay be-
tween the two forces continues through the three states of
aggregation, until cohesion can be ignored in the gaseous state.
It is on account of this latter fact — the absence of cohesion in
gases — that the' gas laws are so simple, and that we base most
of our conclusions regarding the molecular constitution of matter
on the study of gases. So far all this is, of course, not novel.
But Mr. Beilby reduces the whole problem to a question of heat
energy. The mathematician will have to be consulted. But
Mr. Beilby is no doubt right in pointing out that these problems
are more likely to be elucidated by a study of the successive
Crysiallinc and Amorphous States of Metals 425
stages between the absolute zero and the vaporizing tempera-
ture than of the upper ranges where the gaseous state alone pre-
vails. The most instructive field for investigation may be that
of middle temperatures where the opposing forces are more nearly
equal.
The sizes of the ultimate solid particles have been CvStimated
by optical means. Faraday ascribed the various colors exhib-
ited by gold under different conditions to the size of its particles
and their state of aggregation. Ruby glass and ruby solutions,
he proved, are not true solutions, nor molecular diffusions of
gold, but emulsions containing the gold in aggregates of sufficient
size to produce a sensible reflection of light. Zsigmondy and
Siedentopf came to a different conclusion a few years ago. They
succeeded in making these ultra-microscopic particles visible in
the microscope as diffraction disks ; they counted the numbers
of the disks and calculated their size from the intensity of the
reflected light, and as they observed particles ranging from 4 to
791 millionths of a millimeter, they did not believe in any rela-
tion between size and color of the particles. J. Maxwell Garnett
has lately demonstrated that the color of metallic films and
glasses depends not only upon the size of the metallic particles,
but also on the proportion of the volume they occupy in the
medium in which they are diffused.
This is not easy to understand; but Mr. Beilby's observa-
tions, which agree with those of Garnett, suggest an explanation
of much that was puzzling. He thinks that the actual micro-
scopically-visible particles and the larger particles which can be
measured in films, solutions or suspensions, do not in any way
represent the ultimate units of structure which are required by
Garnett, but that they are aggregates of smaller units built up
in more or less open formation. This argument leads us back
to the coarse visible structure with which considerations of the
hard and soft states are concerned.
ABSTRACTS
#
{From recent articles of interest to the Iron and Steel Metallurgist)
OEGREGATION in Steel Ingots. B. Talbot. Paper read at
^ the September, 1905, meeting of the Iron and Steel Insti-
tute. 9,000 w. — The au-
thor reports the results of
an investigation conducted to
ascertain the effect of alu-
minum in decreasing segre-
gation in steel ingots. Both
acid and basic open-hearth
steels were examined, the
weights of the ingots varying
between i^ and 3 J ton.
As a rule, the results show
that in the case of ingots to
which no aluminum has been
added, excessive segregation
down the central line of the
ingot occurs from about 6
inches from the top to about
half-way down the ingot, that sulphur is the element which tends
to segregate most, phosphorus next, followed by carbon, and
finally manganese, the segregation of which latter element is so
slight as to be almost negligible. No silicon determinations were
made, as the amount of this element present was extremely small.
* Note. The publishers will endeavor to supply upon request the full
text of the articles here abstracted, together with all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60
cents, ''D" 80 cents, "E" $1.00, "F" $1.20, '' G " $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cases
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
ing upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.
426
Abstracts
427
The accompanying illustration shows the marked effect
of aluminum in decreasing the segregation of S in an acid open-
hearth steel ingot.
An examination of the results shows clearly that by the
use of aluminum a billet of a much more regular composition
A B
3 c n £
" I I . <r
77-
77!
S. H I
s
Showing Segregation of Sulphur in the Same Heat in the Case of Two
Ingots. A, with Aluminum; B, without Aluminum. Parts
shaded show the areas in which the sulphur has increased 25 per
cent over ladle test. The darker shading shows approximately
areas with over 75 per cent increase of sulphur.
is obtained. This is especially important in the case of carbon,
especially if this steel had without aluminum been intended for
428 The Iron and Steel Magazine
rail purposes, as the surface of the rail would probably have
shown considerable irregularities in the carbon percentage,
with a consequent want of uniformity in its wearing properties.
In cases in which the carbon has segregated to the center,
it is obvious that corresponding areas will be found at the sides
in which the carbon is less than the mean, through the carbon
having migrated to the center to a greater or less extent.
As in cases in which aluminum has been added the segre-
gation is lessened, the distribution of the carbon over the sur-
face of the ingot is found to be much more even, and to approach
more nearly to the composition given by the ladle test.
The author's experience on the addition of aluminum in
the ingot mold during casting has always been that the alumi-
num appears to make the metal set quicker. This, he is aware,
is against the view usually held by metallurgists. Thus, in Mr.
Harbord's recently published book on steel, it is stated: '' The
addition of very small amounts of metallic aluminum to such
metal" {i.e., metal containing dissolved oxides) 'Ms found to
cause a marked increase in the fluidity of the molten metal, to
stop the evolution of gas and to allow of the production of
sound ingots without blowholes." Mr. Harbord, however,
does not seem entirely satisfied with this view, as on the same
page, in referring to Mr. Hadfield's classic work on aluminum
steel, he tells us that, according to that investigator, it is doubt-
ful whether aluminum increases the fluidit}^ of properly made
steel. This latter view falls in with the facts observed by the
author, under whose directions aluminum has been added
regularly to many thousands of casts.
Not only does the addition of a little aluminum to the
metal as it is run into the ingot have a marked effect in setting
the surface, but it also, in the author's experience, tends, when
added above a certain quantity, to form cavities in the apper
part of the ingot, so that the amount added has to be strictly
regulated. The setting effect on the top of the ingot is so
marked that at the works with which the author was connected
after it had become the custom to add aluminum regularly,
the ingots were never sanded over or stoppered down, as no
such treatment was necessary, either with acid or basic open-
hearth steel. With mild steel also it was found that the molds
could be stripped sooner when aluminum had been added.
Abstracts 429
It was also observed that when the same quantity of alu-
minum, viz., some 3 to 4 ounces per ton of steel, was added
to the metal as it ran into the ladle, its effect was not so pro-
nounced as wiien added in the ingot mold as the ingot was
being teemed.
There appears, therefore, to be somewhat of a contradic-
tion in the facts observed, some authorities telling us that alu-
minum in small quantities increases the fluidity of the metal;
others, w^ith whom the author joins, contending that the reverse
is the case. Theoretically, one would undoubtedly expect
some increase in temperature, owing to the reaction between
the dissolved oxides in the metal and the aluminum, an action
akin to the well-known thermite process.
If an increased temperature is obtained, with the conse-
quent increased fluidity, this would cause the steel to take
longer to solidify, and would consequently tend to increase the
segregation, provided that the aluminum has no special action
of its ow^n on the metal, whereas the numerous analyses made
by the author prove that there is a considerable diminution in
the amount of segregation. As the metal appears to set quicker,
and as, consequently, segregation would be expected to be
less, due to this quicker setting, the analyses seem to agree
with this view. The author's usual practice was to add about
3 to 4 ounces of aluminum per ton of steel in the ingot, but
this was never added until the ingot mold was approximately
two thirds full. Assuming the aluminum to be all concen-
trated in this top third of the ingot, it would then only be at
the rate of about 12 ounces per ton, or about .033 per cent
aluminum.
In the author's opinion it would be well worth while for
other investigators, interested in the manufacture of higher
carbon steel, such as for rail, tire and similar purposes, to follow
up these results, with a view to proving whether a more uni-
form and regular steel is not thereby obtained, a result well
worth the few pence per ton the aluminum would cost. Per-
haps the chief result to be looked for would be the decreased
amount of crop end that it would be necessary to cut ofif from
the top of the ingot due to the greater solidity of the top and
the lessened amount of segregation in this top part of the ingot.
This alone would undoubtedly pay for the cost of the alumi-
43© The Iron and Steel Magazine
num added, without considering the more regular quaHty of
the finished product as a whole. No. 419.
I. The Use of Vanadium in Metallurgy, and II. Steel used
for Motor Car Construction in France. Leon Guillet. Paper
read at the September, 1905, meeting of the Iron and Steel In-
stitute. 25,000 w.; illustrated. — The author gives the results
of an extensive research into the properties of vanadium steels.
He concludes as follows:
" It may be afhrmed that in all the cases studied up to the
present vanadium considerably improves the mechanical prop-
erties of metallurgical products. Its effect may be charac-
terized as follows:
'' I. On normal steels it produces a very distinct increase in
the tensile strength and elastic limit, and has no influence, or
an insignificant one only, on elongation and contraction, and
upon resistance to shock. It slightly increases the hardness.
"2. On quenched steels vanadium considerably increases the
tensile strength and elastic limit; it acts in this way with
almost as great an effort as carbon, yet, notwithstanding this,
it does not increase the brittleness.
" The influence of vanadium in metallurgy is thus, in my
opinion, of considerable importance. It is undoubtedly the
element which, together with carbon, acts with the greatest
intensity in the way of improving alloys of iron — that is to say,
in very small percentages.
"It is to be specially noted, however, that allo3^s of iron,
carbon and vanadium are more sensitive to heat treatment and
mechanical handling than ordinary steels, but this does not
appear to be any longer the case in more complex alloys, par-
ticularly in nickel-vanadium steels.
" It remains to consider the influence of the addition of
vanadium upon the cost per cent of the vanadium contained.
The cost of production of ferro -vanadium is such as to readily
allow of its addition, and if, at the moment of writing, the
price of ferro-vanadium is still high (about £1 per pound of
vanadium contained), this must be attributed to the scanty
demand, which is altogether inadequate, and consequently
entails expenses of manufacture which are spread over but a
very small output, thus considerably increasing the price.
Abstracts 43 1
This state of affairs will disappear when the use of vanadium
becomes more widespread. It may be concluded that the
employment of vanadium in the manufacture of special steels
is distinctly indicated, particularly in the manufacture of
quaternary alloys such as iron-nickel-carbon-vanadium.
''It is probable that vanadium will give highly interesting
results with copper and its alloys, but no systematic study
relating to this has yet been carried out."
The author also considers the special steels being used
for motor-car construction in France, which he classifies as
follows :
'' (i) Steels with low percentages of carbon and nickel
(pearlitic steels), which are used for parts which require cement-
ing and quenching, i. e., shafts, gears which engage directly,
etc.
" {2) Steels with medium percentages of carbon and low
percentages of nickel, used, after quenching and reheating, for
a large number of parts, shafts, gearing, pinions, etc.
'' (3) Steels low in carbon and with high percentages of
nickel, used for valves.
'^ (4) Chromium steels, with high carbon and low chromium
percentages, used for bearings.
'' (5) Silicon steels, used for springs and for gearing.
'' (6) Nickel-chromium steels, with low percentages of nickel
and of chromium, employed for numerous parts requiring
resistance to shock, and a certain degree of hardness.
'' (7) A new steel known as NY, the composition of which
has not been published." No. 420.
Methods for the Prevention of Piping in Steel Ingots. (Die
Verfahren zur Verhiitung der Lunkerbildung in Stahlblocken.)
R. M. Daelen. " Stahl und Eisen," August 15, 1905. 1,700 w.,
illustrated. — The author describes the fluid compression and
other methods of preventing piping. No. 421. D.
A Study of the Causes of Blow Holes in a Steel Ingot. (Un-
tersuchung iiber den Ursprung eines Blasenraumes in einem
Flusseisenblocke.) Dr. H. Wedding. " Stahl und Eisen,"
July 15, 1905. 2,400 w., illustrated. No. 422. D.
432
The Iron and Steel Magazine
The Reversible and Irreversible Transformations of Nickel
Steel. L. Dumas. Paper read at the September, 1905, meet-
ing of the Iron and Steel
Institute. 13,000 w. ; illus-
trated.— The author de-
scribes an extensive investi-
gation of the reversible and
irreversible transformations
of nickel steel, from which
he draws the following con-
clusions :
Two facts stand out
above all the others con-
tained in this paper: i.
Nickel manganese and car-
bon introduced into a steel
determine alike the appear-
ance of the same phenome-
non, irreversible transformation, which is more intense the higher
the proportions in which they are present.
2. It is not sufficient that they should be present in the
steel; it is necessary besides, in order that they should exert the
full effects that they are capable of, that they should be in solu-
tion, a state which is often, as regards carbon, impossible of
attainment without the aid of chromium.
Solution. — If it be sought to ascertain what constitutes
the state of solution, it will be found that nickel steels present
the most perfect examples known. Metallography has never
been able to detect in these steels the slightest segregation; no
physical treatment appears capable of destroying their homo-
geneity, and yet they are not chemical compounds of definite
composition. It would seem that chromium nickel steels realize,
in the most perfect manner known, what is termed solid solu-
tion.
Saline solutions are subjected to internal stresses, called
osmotic pressure; it would not be surprising, therefore, that an
internal stress of growing intensity should be occasioned in the
steels in proportion as nickel is added; it reveals itself by an
internal " working," resembling that of steels which have under-
gone the irreversible transformation.
Abstracts 433
Allot ro pic Transformations of Iron. — Why is it that
the internal stresses disappear so suddenly when the added
elements reach a certain proportion? In order to answer this
question it is necessary to bring into consideration a very re-
markable property of iron; that of being able to undergo allo-
tropic modifications.
We may say, therefore, in accordance with the terminology
adopted by Mr. Osmond, that the state of internal tension
disappears when the point Ar3, below which iron ceases to exist
in the y state, is lowered, by means of a proportion of nickel,
or of other elements, to below the ordinary temperature. Now
when the iron is free from all alloys, this point is situated at
850° C. It is the dissolved elements which have retarded the
transformation; this may well be admitted, inasmuch as the
point Ar3 is seen to fall gradually in proportion as the additions
increase.
So long as the iron is in the y state in the steel, no internal
stresses arise. A stress, even when severe, may be the result
of an external force. This is a fact which I have shown to be
related to the other fact that the volume increases when the
transformation manifests itself by an internal working. The
appearance of the internal stresses is one of the characteris-
tic manifestations of /? iron produced by the transformation,
which is also shown by the microscope, by the martensitic struc-
ture, and in the mechanical tests by a considerable increase in
the hardness and brittleness. Nickel steels with high pro-
portions of /? iron are those which contain large amounts of
foreign elements, and have undergone the irreversible trans-
formation. The transformation also causes a portion of the
iron to pass into the a state; it is this portion alone which
is magnetic. It is at its maximum proportion in unworked
steels which contain the minimum proportions of foreign ele-
ments.
These considerations will permit of the whole of the com-
plex facts related in the foregoing account being still more suc-
cinctly stated.
The properties of nickel steel are the same as those of iron,
modified, and even exaggerated, by the influence of the elements
it holds in solution, yet at the same time perfectly recognizable,
particularly that of undergoing alio tropic modifications.
^24 The Iron and Steel Magazine
The movement of the molecules is hampered by the presence
of nickel or of other elements. Mr. Hadfield has recently given
a peculiarly striking example of this by submitting to mechanical
tests in liquid air Swedish irons of great purity and nickel steels.
At the temperature of liquid air the Swedish iron became ex-
tremely brittle, but it recovered all its toughness on returning to
the ordinarv temperature. At the same temperature the nickel
steel had lost very little of its tenacity, because the nickel im-
peded the movement of the molecules, and prevented them from
forming crystals.
The highly remarkable properties of those nickel steels
which have not undergone the irreversible transformation have
occupied the principal place in the foregoing paper. Because of
their high percentages of nickel they are, unfortunately, so
costly to produce that they are beyond the range of current
industrial apphcation. It would seem, therefore, at first sight,
that a minute study of their properties was almost useless from
a practical point of view.
The direct utiUty of these investigations is, perhaps, greater
than might be apparent at first sight, but I will confine myself
to remarking that they possess an indirect utihty of prime
importance, in showing the nature of the operation involved in
adding nickel to steel.
Two effects are produced: the homogeneity is increased,
and a form of what is otherwise known as '' working " results;
the condition of mutual solution is improved, and /? iron is
produced. With very low percentages the first effect pre-
dominates, the crystaUization of the iron is impeded and the
brittleness lessened. The action of the nickel is analogous to
that of a quenching carried out at looo degrees, which, as is
known, greatly diminishes the brittleness of soft steels. It may
be remarked, in passing, that it is much more practicable to
introduce nickel than to quench when large masses are concerned.
The second effect, intensification of the proportion of /? iron,
which counterbalances the first named, must not, however, be
lost sight of. The improvement in the state of reciprocal
solution impedes crystallization, but occasions osmotic pressure,
i.e., both removes and confers brittleness. This explains why
additions of nickel, which, up to about 2 per cent, are devoid of
danger, begin to become so above this percentage. The steel
.4 hstracts
435
early assumes the characteristics of a highly worked metal,
corresponding to a quenched carbon steel. Above 8 or lo
per cent of nickel the steel is difficult of employment, that is,
until the proportion reaches that which induces the transition of
the iron to the ;- state.
Will the application of steels of this category always remain
somewhat rare? It may be permitted to doubt it, and the suc-
cess of Mr. Hadfield's non-magnetic manganese steel justifies this
doubt. In the meanwhile there is nothing to hinder the extended
use of steels with low percentages of nickel ; they are admirably
adapted in all cases where it is important to diminish brittle-
ness or to reduce weight. No. 423.
The Nature of Troostite.
Carl Benedicks. Paper pre-
sented at the September,
1905, meeting of the Iron and
Steel Institute. 5,000 w., il-
lustrated. — The author re-
calls the original definitions of
Osmond regarding the nature
of troostite and opposes Boyn-
ton's suggestion that troostite
might be ^ iron. ^^ It is
quite natural to assume," the
author writes, '' that troostite
is that part of martensite in
which cementite has just be-
gun to form (incipient pearlite
formation), but that owing to the rapidity of the cooling, the
separate particles of cementite do not attain to such a size
that they can be distinguished microscopically."
Boynton's theory must fall or stand by the presence or
absence of troostite in hyper-eutectoid steel, and the author
contends that Boynton avoids the difficulty by calling this
troostite '' sorbite." ^' The facts," he writes, '^ are in entire
accord with the hypothesis that troostite is a pearlite with ultra-
microscopically small particles of cementite." The author,
moreover, is inclined to consider troostite as resulting from a
transformation in situ of martensite, although Osmond did not
436 The Iron and Steel Magazine
wish to insist upon this. The following conclusions represent
the author's views.
In confonnity with this the author has already formulated in
the work referred to the difference that can appropriately be made
between troostite and sorbite, and will venture to repeat it here:
'' If a piece of martensite or austenite has already begun to
undergo transformation which ultimately results in pearlite, we
have troostite.
'' If the martensite has not succeeded in completely resolv-
ing itself into pearlite, we have sorbite.
'^ Troostite is the first, sorbite the last, stage of transforma-
tion between martensite and pearlite."
In the author's opinion one might set up this practical
criterion: If the bulk is composed of martensite (± austenite),
but certain parts become dark on etching, these are called
troostite; if the bulk is composed of pearlite, but certain parts
become dark on etching, these are called sorbite. But this
question is and remains one of names.
In conformity with the experimental works of Osraond,
Le Chatelier, Heyn, Boynton and Kourbatoff and with his own
observations, the author is led to the following conclusions:
1. Everything we know at present points to the fact that
Osmond has quite correctly defined troostite as an intermediate
form between martensite and pearlite.
2. Between troostite and pearlite there is a continuous
transition, and one is naturally led to the conclusion that troostite
is a pearlite with ultra-microscopically small particles of cemen-
tite (containing also more or less hardening carbon). This
hypothesis is the simplest and the most natural one to accept
until it be shown that it is contrary to truth.
This is in keeping with the fact of the extreme ease with
which troostite is affected by reagents, with its varying hard-
ness, which lies between that of martensite and that of pearlite,
and with the fact that troostite, from a steel rich in carbon,
yields cementite when slightly heated, as Kourbatoff has shown.
3. In all probability troostite is formed by a transformation
in situ of martensite, which implies that the carbon it contains
must be the same as in the martensite from which it originates.
Boynton's assumption that troostite is pure ^ iron is with-
out experimental or theoretical support. Similarly Kourbatoft"'s
'Abstracts 437
suggested theory, that troostite is a solution of elementary
carbon in iron, is tmtenable.
4. Troostite is formed from martcnsite by the appropriate
lessening of the intensity of the hardening, especially in places
that are in contact with ferrite or cementite; this is in perfect
conformity to theoretic requirements.
5. As far as we can judge, troostite offers in the domain
of alloys an interesting analogy to the colloid solutions.
Since the above was written, Rogers, in a paper on troostite,
has taken into consideration the different arguments in favor
of troostite containing carbon, contrary to Boynton's hypothesis.
He says, however, that he fails to see why troostite should be /?
iron rather than ;'. No. 424.
The Influence of Nickel and Carbon on Iron. G. B. Water-
house. Paper read at the September, 1905, meeting of the Iron
and Steel Institute. 12,000 w., illustrated. — The author
describes the results of an investigation conducted in the Metal-
lurgical Laboratory of the School of Mines of Columbia Univer-
sity and consisting in the study of the properties of a series of
steel of constant nickel content (about 3.80%) with varying
carbon percentages, while the other elements were kept as low
and constant as possible. The following conclusions are drawn
from the results obtained:
1. Nickel decidedly raises the tenacity without materially
lowering the ductility. The elastic ratio in pure nickel-carbon
steels is only slightly greater than that of carbon steels.
2. Annealing has a marked influence; it lowers the tenacity
without greatly raising the ductility.
3. The constituents of steels with low-percentage nickel
in the unquenched state are; ferrite, pearlite, cementite and
graphitic carbon.
4. The pearlite of these steels shows a great readiness to
segregate into its constituents: ferrite and cementite.
5 . In this condition the cementite has the formula Fe (Ni)3 C.
6. The eutectoid ratio in these steels appears to lie at about
.70 per cent carbon, but in the rolled steels no free cementite
show^s until the carbon reaches about i.oo per cent.
7. Nickel lowers the transformation points Ar3.2 and An
about 20 degrees for every i per cent of nickel.
438 The Iron and Steel Magazine
8. The cementite of these steels is very Hable to precipitate
its carbon as '^ temper graphite." No. 425.
Process for Converting Fine Iron Ores into Nodules. '' Iron
Age," September 7, 1905. 2,500 w., illustrated. — Description of
a plant of the National Metallurgic Company at Newark Bay,
N. J., for the production of ore nodules from fine ores. The
process is being applied chiefly to the residue from the treatment
of iron pyrites for the production of sulphuric acid and which is
known as ^' blue billy." The present process differs from
previous ones in that it employs tar (though patent claims are
made also on other adhesive substances and carbohydrate com-
pounds) which has an affinity for and forms volatile compounds
with such impurities as sulphur and arsenic. Along with these
impurities it is gradually volatilized, the iron oxides being con-
verted into nodules of any desired size, free from moisture. The
adhesive substance is not used to bind the particles of ore to-
gether permanently, but its function is to bind them initially,
and in the progress of the ore through the rotary kiln (see illus-
tration) used for the production of nodules aid in fusion, the
final product being permanently coherent nodules containing
substantially no fixed foreign compounds.
It is found that the size of the nodules can be regulated by
varying the quantity and quality of the binder, the degree of
heat and the rapidity of movement of the ore through the kiln.
In practice under Mr. King's process the addition of i per cent of
pitch to 99 per cent of an ore analyzing 67 per cent metallic iron
and I per cent silica produces nodules about the size of a goose
egg, a size adapted to open-hearth work, while ^ per cent of
pitch added to 99^ per cent of the same ore produces nodules the
size of a partridge egg, the size best adapted to blast-furnace
practice.
The National Metallurgic Company has erected what it
terms an ore purifying and nodulizing plant, with a capacity of
200 tons in twenty -four hours. The object has been to construct
as far as possible an automatic or continuously operating plant.
The ore, or pyritic cinder, is received by railroad or boat, is
stored under a trestle and in bins, and from the bins or from the
stock piles is mechanically conveyed to a rotary kiln for treat-
ment.
Abstracts
439
Referring to the elevation of the cinder treating plant, it
will be seen that the cars drop the ore into a series of bins under
the trestle. These bins discharge either directly into a crusher or,
if the ore is in such a condition of fineness as not to require
crushing, into a storage tank located above the feed end of
the rotary kiln. From this storage tank the material is con-
veyed by screw conveyor to a feed pipe which projects down and
into the end of the kiln. At the same time that the fine ore is
delivered from the storage tank tar is dropped into it as it
reaches the end of the screw conveyor farthest from the tank.
There is thus a thorough mixing prior to the admission of the ore
into the kiln. The tar tank surrounds the exhaust chimney of
the kiln, and thus the tar is kept warm and fluid. Held in
masses by the preliminary binding of the tar the ore moves along
Elevation of Ore Storage and Feed Bin, Rotary Kiln and Nodule Conveying
and Storage Plant
in the kiln under a horizontally rotary motion, encountering
different degrees of heat, the temperature increasing toward the
farther end of the kiln. Before merging from the kiln the
nodules are agglomerated or permanently semifused ; progressing
still further they are discharged through a hood and conveyed
by elevators to the nodule tank for final disposition.
The fusion of the ore is accomplished by the injection of
powdered coal into the discharge end of the kiln, and an impor-
tant feature of the plant is the coal dryer and its accessories.
The dr\^er is located adjacent to the bins under the trestle, and
the coal is carried by conveyors to the dryer bins and through the
latter automatically. The dr5^er deviates from the usual form
in that the heated gases and waste products of combustion after
passing on the outside of the dryer return to the front, and thence
440 The Iron and Steel Magazine
pass directly over the coal dryer to discharge through the chim-
ney stack.
From the dr3^er the coal is discharged into a bin from which
it is conveyed to a Griffith mill. The latter runs at i,8oo revolu-
tions and produces a ground coal of which 90 per cent will pass a
loo-mesh sieve. A screw conveyor brings it to the ground
coal tank, thence it is propelled by screw conveyor and dropped
into a blow pipe, a fan blowing it into the hood at the discharge
end of the rotary kiln. The point of fusion in the kiln depends
on the amount of air pressure, the fusion zone being either drawn
toward the hood or thrust further back toward the feed end, as
required by the desired size of ore nodules.
The field of the process described above is broader than
the operations thus far carried on. Its use for the conversion of
fiue dust into nodules is under consideration, the flue-dust
problem having assumed no small proportions under the increas-
ing use of Mesabi ores. But a more important field is capable of
development by the installation of ore-nodulizing plants at Lake
Superior iron mines, whose ores are of unusual fineness. The
removal of moisture from Lake Superior ores previous to their
shipment from the mine has been agitated for a number of
years. It has been considered a commercial proposition to dry
these ores at the mine, but prosperity has pushed any serious
operation in this direction into the future. A process that
removes moisture and at the same time converts the ore into
such form that practically no flue dust is produced is naturally
important to Lake Superior mining interests, as well as to fur-
nacemen using Lake ores, and it may be stated that some steps
are in contemplation for such an application of the process above
described. No. 426. A.
Note on the Occurrence of Capper, Cobalt and Nickel in
American Pig Iron. E. D. Campbell. Paper presented at the
September, 1905, meeting of the Iron and Steel Institute. 1,800
w. ■ — The author reports the results of the analysis of many
samples of American pig iron for nickel cobalt and nickel, and he
finds that two points of interest are brought out by a study of
these results.
First, the fact that two samples, made from ores from the
Lake Superior district, contain no copper, cobalt or nickel,
.4 bstracts 44 1
although the Gogebic Range is comparatively near the great
copper-producing district of Lake Superior. The second point
relates to samples which are the only ones containing any con-
siderable amount of cobalt or nickel, and both of these irons have
gained a reputation for their valuable properties for car-wheel
castings. The author has no information in regard to the influ-
ence of small amounts of cobalt or nickel on the properties of
cast iron, and the occurrence of these elements in these two
irons mentioned may be a mere coincidence, but it is at least
suggestive. No. 427.
A Manipulator for Steel Bars. Douglas Upton. Paper read
at the September, 1905, meeting of the Iron and Steel Institute.
1,000 w., illustrated. — The author describes a manipulator for
steel bars which consists of a series of three machines, the first
of which consists of two pairs of movable heads which are installed
at each side of the roughing and finishing rolls. These heads can
move in either direction parallel to the rolls. They are prefer-
ably actuated by hydraulic cylinders, and they place the piece
in position for entering any required pass. These heads, more-
over, are fitted with levers for turning the piece up for edging
purposes; not only will they turn the piece up, but without
actually gripping it, will maintain it in that position until it
enters the rolls. No. 428.
The Keep Sectional Cupola. ''The Iron Age," August 31,
1905. 1,400 w., illustrated. — The article describes a recent
cupola furnace designed by W. J. Keep and built by the Northern
Engineering Works, Detroit, Mich. The furnace is composed of
four sections easily handled and assembled. No. 429. A.
Car Wheel Forging. James H. Baker. '' The Iron Age,"
September 7, 1905. 2,200 w., illustrated. — The author de-
scribes the forging of car wheels as conducted by the Solid Steel
Tool and Forge Company, Pittsburg, Pa. No. 430. A.
A Process for the Prevention of Piping in Large Steel Ingots.
(Verfahren zur Verhiitung der Lunkerbildung in Schweren
Rohstahlblocken.) F. O. Beikirch. " Stahlund Eisen," August
I, 1905. 1,800 w., illustrated. — The process consists in the use
442 The Iron and Steel Magazine
of sinking heads which are afterwards removed, together with the
piped portion. No. 431. D.
The Department of Iron and Steel Metallurgy at the Uni-
versity of Sheffield. J. A. Arnold. Paper read at the September,
1905, meeting of the Iron and Steel Institute. 2,800 w., illus-
trated. No. 432.
Cleaning Blast-Furnace Gas. Axel Sahhn. ^' Cassier's
Magazine," October, 1905. 5,500 w., illustrated.— No. 433- B.
METALLURGICAL NOTES AND COMMENTS
Robert Forrester Robert Forrester Mushet (see frontispiece)
Mushet was the son of David Mushet, the discov-
erer of the Scotch Black Band iron stone, and like his father
an enthusiastic metallurgist. His most important achievements
are the invention of self-hardening steel and the addition of
spiegeleisen to the refined metal at the end of the Bessemer blow.
In a recent issue of " The Ironmonger," the advent of self-
hardening steel is briefly related in the following words :
" Like other important inventions, that of self -hardening
tool steel was an accidental discovery, due to the observation of
the effect of a draught of air upon some tools which had been
left near the bottom of a door. It was then proved that if tools
were reheated to a full scaling or almost yellow heat their quality
was improved. Mr. Mushet himself, while experimenting for an-
other purpose, found that one of his trial bars had the property
of becoming hard, after being heated, without the hitherto
necessary water-quenching, and that this property was due to the
presence of tungsten. In 187 1, the manufacture of Mushet steel
was taken up by S. Osborn & Co., of the Clyde Steelworks,
Sheffield, and its superior cutting qualities soon gained it a world
wide reputation. Subsequently the majority of the Sheffield
steelmakers added a self -hardening tool steel to their products.
Mushet steel was undoubtedly a forerunner of the modern high-
speed steel."
The part taken by Mushet in the final success of Bessemer's
pneumatic process is too well known to need repetition in these
columns. While it would be unwarranted to afhrm that were it
not for his cooperation the Bessemer process would have failed,
it may at least be said that his assistance greatly hastened the
success of the process. It is probable that the following remarks
of William Menelaus, then president of the Iron and Steel Insti-
tute, on the occasion of the presentation to Mushet of the
Bessemer Gold Medal in 1876, correctly express the opinion of
metallurgists at that time and that the same opinion still prevails
443
444 The Iron and Steel Magazine
to-day. '^ It is needless to inquire very particularly what
success attended Mr. Mushet's attempts to improve old processes,
because they were all overshadowed by the beautiful invention
of the spiegeleisen process, as applied to his friend Mr. Bessemer 's
great invention, and it was upon that ground that the council
resolved to pay Mr. Mushet the compliment that they then did.
He thought they would agree with him that the application of
spiegeleisen, in the way it was done to Mr. Bessemer 's invention,
was one of the most elegant, as it certainly was one of the most
useful, inventions ever made in the whole history of metallurgy,
and he thought it would be conceded also that for that alone, if
for nothing else, Mr. Mushet well deserved the compliment they
were about to pay him. It was an invention which was worthy
of being associated with the great invention of Mr. Bessemer.
The two inventions would go down together; in fact, the one
was the complement of the other, and he thought he was right in
saying that no man in that room could be better pleased than his
friend, Mr. Bessemer, that the council had resolved to pay that
compliment to Mr. Mushet. He thought it was a fit and seemly
thing that the medal instituted by Mr. Bessemer should com-
pliment the man who had made what he thought was really a
brilliant invention. It had made the invention of Mr. Bessemer
perfect, and probably would be used in England as long as
Bessemer metal was made."
It is well known that Mr. Bessemer never recognized the val-
idity of Mushet's patents, and this is forcibly stated in his recent
autobiography, from which the following remarks are quoted:
" For a period of more than two and a half years (1857-60)
after the date of Mr. Mushet's three manganese patents, I had no
intimation of any kind that either I, or my licensees, were in-
fringing any of these patents. But about three or four months
prior to the date when a further ;^ioo stamp was required to be
impressed on them, to prevent their forfeittire, I received a letter
from a Mr. Clare, of Birmingham, calling himself Mr. Mushet's
agent for the sale of steel, and requesting an interview with me
and my partner at my office in London on the following morning.
On his arrival, he explained the object of his visit; it was simply
to say that Mr. Mushet was prepared to grant me a license to use
his manganese patents for a nominal sum ; he merely wanted his
rights acknowledged. I then told Mr. Clare that we considered
Metallurgical Notes and Comments 445
that Mr. Mushet had acquired no rights under either of his three
manganese patents, and that we entirely repudiated them. I
also told him that we were anxious to meet any claims legally
preferred; that we were prepared, on any day to be mutually
arranged, to receive Mr. Mushet and his solicitors and witnesses
at the Sheffield works ; that we would allow them to see the crude
iron converted and recarburized with spiegeleisen, made into an
ingot, and forged into a bar, and that I would personally take that
bar to one of my customers and sell it to him in their presence ;
and then the prosecution of our firm for infringement would be a
very simple matter. This ofTer resulted in Mr. Clare's retirement
from my office, and after that interview we never heard from him,
or from Mr. Mushet, on the subject."
Mr. Bessemer also writes that he paid Mushet over ;^7,ooo
in annual allowances of £300 and other payments. The reason
for these disbursements Mr. Bessemer also states in his auto-
biography.
'' There was a strong desire on my part to make him my
debtor rather than the reverse, and the payment had other ad-
vantages; the press at that time was violently attacking my
patent, and there was the chance that if any of my licensees were
thus induced to resist my claims, all the rest might follow the
example, and these large monthly payments might cease for such
a period as the contest in the law courts might last. The annoy-
ance, if nothing else, would have been very great, and I had
neither time nor patience to wage a paper war from year's end
to year's end with unscrupulous writers. In the hope that an
allowance to Mr. Mushet might have the effect of restraining
these attacks on me, I offered to pay him ;^300 a year, aiming at
abating an intolerable nuisance which I had no other means of
preventing. While we were paying over £3,000 per annum in
the form of income tax, the £300 was but a small additional tax
on my resources, so I allowed it to drag on until Mr. Mushet's
decease, in 1891, having thus paid him over £7,000. So, nat-
urally, ends this part of this history of my invention, as far as
Mr. Mushet is concerned." .
Electric Smelting.* — With the development of electro-
metallurgical processes in other parts of the world the possibility
* " Engineering Magazine," October, 1905.
446 The Iron and Steel Magazine
of applying the methods of electric smelting of iron and refining
of steel to the industry in Great Britain is attracting attention.
In a paper recently presented before the Manchester section of
the Society of Chemical Industry, Mr. R. S. Hutton shows that the
position of England in this respect is by no means so discouraging
as some would believe. Incredulity in this respect is largely
based upon the fact of the limited water power available, but
it must not be forgotten that the power cost is but one element,
and not always the controlling one.
'^ Many leading electrical engineers have published minute
data as to the cost of power generation, but only for electric
lighting, traction and motor purposes. These cases, however,
are so entirely different in character from those we are concerned
with that no definite conclusions can be drawn from this evidence.
The average power station is fortunate to get a 15 per cent load
factor, and overjoyed with 30 per cent; whereas, in nearly all
electro-chemical works, the manufacture is continued night and
day throughout the year, and the load factor may be taken as
100 per cent. If one may judge from the evidence of those few
who have had actual experience in electro-chemical industries
using steam power, a figure of £6 to ;^8 for the horse-power year
is quite attainable under such conditions. With producer gas
this may probably already be brought down to £4. in Great
Britain. For the sake of comparison, it may be pointed out that
although in some few places in the Alps and Norway a figure as
low as 175. per horse-power year has been attained, water power
is very seldom to be obtained so cheaply. At Niagara the price
of supply to large consumers varies from £t^ 11s. to £/\. t,s. and at
Rheinfelden reaches £6 for the horse-power year.
'' So far as blast-furnace gas is concerned, no very sure data
are available for similar industries. The supply of cheap gas
power is likely to prove so beneficial to Great Britain in the ap-
plication of the industries we are about to consider that it is ear-
nestly to be hoped that those who are concerned in the construc-
tion of large gas engines will be led to take an interest in these
developments. With their cooperation the number of remunera-
tive electro-chemical industries may be very largely increased in
England."
There does not appear to be any immediate probability of
the introduction of electric methods for the reduction of iron
Mcialluri^ical Notes inid Comments 447
from the ore, but Mr. Mutton gives some interesting points about
the relations of electro-thermic processes to the present smelting
methods.
" The application of the electric furnace to the metallurgy
of iron, with the exception of some few small-scale experiments,
which are more of historical than technical interest, may be said
to be largely founded on the experience gained in the manu-
facture of calcium carbide. Carbide furnaces have been and are
being largely used for the production of rich ferro-alloys such as
ferro -chromium and ferro-silicon, and in this way electro-metal-
lurgv has already been of considerable service to the steel in-
dustry. As the production of calcium carbide became less and
less remunerative, and as the demands of the market for these
ferro-alloys became satisfied, definite attempts were made to
tackle the problem of the direct reduction of iron ores.
'' It might seem to be rather a hopeless task which the
electro-metallurgist has thus set himself; for direct competition
w4th the blast furnace is obviously out of the question so far as
our own and probably all other present iron-producing countries
are concerned. On the other hand, there are certain advantages
which can be gained by electric heating, and although the electric
reduction of iron ores is at the moment unremunerative, we may
expect to hear more of it in the future, when the general devel-
opment of electric furnace construction is more advanced.
'' Every ton of pig iron produced in the blast furnace
requires very nearly one ton of coke for its production. Of this
amount only one third is necessary for the chemical reduction of
the ore, the balance being employed in producing and maintain-
ing the requisite temperature. This two thirds of the fuel supply
can be replaced by electric heating.
" In actual practice, so far, only the simple case of reducing
the ore and allowing the carbon monoxide to pass away unused
has been tried. Various methods have, however, been proposed
by Heroult, Harmet and others for utilizing the total heat of
combustion of carbon. Under these conditions it should be
possiVjle to reduce iron ore with a much smaller power expendi-
ture. The perfecting of methods along these lines is a matter
for the future."
In considering the question of the electric production of
steel the matter should be examined on its own merits, and not
448 The Iron and Steel Magazine
in connection with the ordinary processes. So far as the matter
of power is concerned the results of the investigations of the
Canadian commission show the power expenditure per 1,000-
kilograms of steel to range between 500 and t,ooo kilowatt hours,
the lowest expenditure being that of the Kjellin induction
furnace, which was charged with molten pig and cold scraps
while the largest power expenditure occurred when the charge
was cold.
More important than the power consumption, however, is
the quality of the steel produced by the employment of electric
refining processes.
'^ In the first place, there seems to be good evidence to show
that steel, equal in quality to the best Sheffield crucible steels,
can be produced in the electric furnace. This can be accom-
plished either in such a furnace as that of Kjellin which, in its
present form, is used almost entirely for melting up carefully -
chosen raw materials, and does not rely on any considerable
refining of the material. On the other hand, with the Heroult
furnace such a product can be obtained starting with almost any
grade of raw material; this process relying essentially on its
capability of rapidly and completely refining pig iron or ordinary
scrap steel. The economical advantages of using a cheap grade
of raw material for producing high quality crucible steel will
doubtless tell in favor of such a method. It is largely to such
possibilities of refining and to the relatively high cost of fuel
per ton of product for the manufacture of crucible steel that the
electrical processes owe their advantages.
'' From the investigations of the Canadian commission it
appears that in nearly all cases the whole operation of meltings
and refining the raw material has been effected by electric heat-
ing. In Great Britain, where coal is cheap, it is almost certain
that much of this heating could be more economically carried
out by the combustion of fuel. In the case where molten iron or
low-grade steel can be run into the electric furnace, it will be seen
that the power expenditure required for refining it and bringing^
up its quality to that of a crucible steel is indeed very low.
Along such lines as these, the electric furnace may be expected
to find still wider application than to the manufacture of high-
grade crucible steel."
Metallurgical Notes and Comments 449
The Metallurgical Congress at Li^ge.* - — We have given else-
where in this issue a fully illustrated account of the engineering
features of the international exposition now being held in Li6ge,
Belgium, and, following the custom usual with such exhibitions
there have been held various technical congresses and scientific
gatherings, concerning which reports are now beginning to appear.
In recent issues of *' Le Genie Civil " is given an excellent sum-
mary of the proceedings of the Congress of Metallurgy from
the pen of the well-known metallurgist, M. Leon Guillet, himself
an active contributor to the work of the congress.
The congress, which was attended by about 1,600 delegates,
including some of the most eminent metallurgists and engineers
from Belgium, France, Germany, England, etc., had its work
divided into five departments, treating of groups of metallur-
gical work. The first section considered problems relating to
large operations, such as the manufacture of pig iron and of
steel, the production of power from furnace gases, the utilization
of slag, etc. The work of the second section related to special
methods and progresses, such as the manufacture of alloy steels,
electro-metallurgical processes and the like. In the third
section there was discussed the treatment of iron and steel
products, such as the heat treatment of steel, and the effects
of rolling and mechanical working of the metal. The fourth
section was devoted to the study of metallography, while the
fifth took up processes having an indirect relation to metallurgy,
such as the brazing and welding of metals, etc.
It is impracticable, within the space here available, to com-
ment upon all the papers submitted before the congress, but
some of the more important communications may be noticed.
Thus M. E. Bian, the director of the iron works at Eich, in Lux-
emburg, described the method which he has found satisfactory
for purifying the gases from blast furnaces, rendering them
suitable for use in gas engines. This apparatus consists of a
cylindrical chamber filled half full of water, and containing a
shaft carrying a number of disks of metallic netting, these disks
being kept in constant rotation. The gas, passing through these,
parts with its dust and other matter in suspension, and the disks
are continually washing themselves in the water, this latter
being constantly renewed.
* " The Engineering Magazine," October, 1905.
450 The Iron and Steel Magazine
An interesting matter in connection with the work of the
first section was the discussion of the Gayley dry-air blast for
blast furnaces. The correctness of the theory of the process is
fully borne out by the facts developed in the discussion. Thus
a table prepared by M. Divary, of the Creusot works, shows that
the fuel consumption of a furnace under his observation bore a
close relation to the hygrometric condition of the atmosphere.
Taking the fuel consumption in January as a base, there being
6.3 grams of water per cubic meter of air, and the daily pro-
duction of the furnace being 90 metric tons, there appeared in
July, when a cubic meter of air contained 13.6 grams of water,
an excess consumption of 133 kilograms of coke, while the out-
put of the furnace fell to 70 tons per day. The results for other
months showed a close correspondence as to the variation in
coke consumption and in iron production, with the variation in
the proportion of moisture in the air, these figures agreeing
closely with those observed by Mr. Gayley at Pittsburg. It is
intended to introduce the refrigerating process of drying the air
at the Creusot works, as well as in other establishments in Bel-
gium and Germany.
In discussing the theory of the dry-air blast, M. Le Chatelier
showed the injurious action of moisture in the air in connection
with the presence of sulphur in the iron. By the use of air which
is free from moisture any sulphur which is present is converted
into sulphurous anhydride, which is entirely absorbed by the
limestone in the upper zones of the furnace, where iron itself
has not yet reached the spongy condition in which it can take
up the gas. The sulphur thus passes off entirely in the slag, a
condition which does not occur in the presence of moisture.
The much disputed subject of slag cements came up for
discussion at the congress, and Professor Wedding, of Berlin,
expressed himself of opinion that slag Portland cement, made
by recalcining and grinding briquettes made of granulated slag
and lime, does not differ chemically from ordinary Portland
cement, while the results of mechanical tests are entirely com-
parable.
Passing to the work of the second section, this included dis-
cussions upon special alloy steels, M. Guillet himself furnishing
a classification of these products, according to the manner in
which the added metal combines with the iron and with the
Metallurgical Notes and Comments 451
carbon. M. Guillet gave several diagrams showing the influence
of the difterent constituents upon the properties of resistance to
rupture, to shock, to elongation and upon hardness, these
enabling a general idea of the effects of the various constituents
to be determined and compared. Referring to the ternary
steels, these including those containing iron, carbon and one
other constituent, M. Guillet says that the nickel and the man-
ganese steels may sometimes take the place of carbon steels;
that the polyhedric steels should have an important industrial
future if the price can be brought sufficiently low; that the
tungsten and molybdenum steels have shown themselves of
great value for high-speed tools ; and that there is apparently no
practical use for the graphite steels.
So far as the interesting subject of electrometallurgy is con-
cerned, the principal point brought out at the congress was the
fact that the well-known projectile works of Jacob Holtzer, at
Unieux, has put into service a steel-refining furnace of 1,000
kilowatts, capable of producing 7,000 to 8,000 kilograms of steel
at a charge.
In connection with the works of the section devoted to
processes of treatment of iron and steel, mention may be made
of the researches of Hadfield upon the effects of low temperatures
upon alloy steels, already noticed in these columns; and of the
paper of M. Creplet, upon the application of electric power to the
driving of rolling mills.
Of especial importance was the paper of M. Le Chatelier, in
the fourth section, upon the subject of metals and alloys by the
methods of metallography. M. Le Chatelier discusses the
methods of polishing the surface of the metal to avoid surface
hardening, describing the preparation of the emery and alumina
for working the surface, and the use of various solutions for
etching the polished metal. The use of picric acid, originating
in the laboratory of M. Le Chatelier, is now well known, but a
later method in the use of heated saline solutions containing
an oxidizing substance. By using a 25 per cent solution of
caustic soda with 2 per cent of picric acid, heated to 100° C,
the cementite is attacked, without any action being produced
upon the other constituents.
In the production of the microphotographs M. Le Chatelier
prefers the Nernst lamp to the mercury arc, and the details of
452 The Iron and Steel Magazine
his microscopic apparatus have been worked out with great care.
After all the care which can be taken, much depends upon the
skill and judgment of the operator, and in this, as in other depart-
ments of investigation, it is impossible to be assured of uniform
results.
In the auxiliary metallurgical subjects attention was
directed at the congress to the use of the oxy -hydrogen and the
oxy-acetylene blow-pipes for the welding of metals. There ap-
pears to be no doubt that satisfactory welds may be made with
either apparatus, and the decision from the industrial viewpoint
depends mainly upon the cost. The use of electrolytically
produced gases is of interest so far as the oxy -hydrogen apparatus
is concerned, but the advantage in point of cheapness appears to
lie with the oxy-acetylene blowpipe.
The Cause of Brittleness in Steel.* — Among the various
physical and mechanical properties of the numerous iron alloys
grouped under the generic name of steel, that of brittleness, or
as the French call it, fragility, has caused much perplexity. Two
products, apparentlv the same in chemical composition and
in visual constitution, will be found to be quite different in
behavior under shock, one being tough and resistant, while the
other breaks without warning. Attempts to devise physical
tests to discover the causes of this action, or at least to separate
the good material from the bad, have met with but partial
success, although various forms of drop tests are now realized to
be of much value in commercial investigations. Examinations
of portions of structures which have failed in service have not
given any very clear indications as to the causes of sudden
breakages, and it has been felt that some better knowledge of the
origin of the property of brittleness would have to be discovered
before the practical side of the subject could be pursued further
to any material advantage.
We now have a paper in the '' Revue de M^tallurgie," by
M. Hjalmar Braune, giving a theory of brittleness based upon
his investigations of the past six years, this paper being of a
preliminary nature and to be followed by a more detailed account
of the experimental researches upon which it is based.
Briefly, M. Braune maintains that brittleness is due to the
* " Engineering Magazine," October, 1905.
Metallurgical Xotcs and Comments 453
presence of combined nitrogen taken up by the iron during
various stages of its manufacture. The nitrogen appears to be
combined entirely with the iron itself, with the ferrite, forming
what may be called a nitrogenized iron ; the carbides such as the
cementite being entirely free from any nitrogen. This nitride
of iron appears as a solid in solution in the ferrite, and acts to
lower its point of fusion and at the same time diminishes its
capacity for dissolving carbides of iron. In this way the nitrogen
exerts a marked influence upon the quality of the metal, whether
it be a soft iron or a hard steel or cast iron.
These statements may be proved by a few simple experi-
ments. A test piece of iron or steel of the highest quality is
placed in an atmosphere of ammonia, and raised to a temperature
of 800° C. for a period more or less prolonged. These pieces
are then annealed in sand, to permit the combined nitrogen to
become homogeneously distributed through the metal. The
behavior of pieces thus treated, when tested, shows very clearly
the influence of the nitrogen upon the resistance. When the
content of nitrogen reaches 0.07 to 0.08 per cent the elongation
rapidly diminishes, and becomes discontinuous, while for higher
percentages of nitrogen the ductility of the metal practically
disappears. In some instances the surfaces of the test pieces
became covered with fine cracks, these effects appearing in
pieces in which the annealing had not been sufficiently prolonged,
so that the greater portion of the nitrogen remained near the
surface.
The effect of the presence of the nitrogen may also be ob-
served by making a metallographical examination of the test
specimens. The original untreated metal showed a constitution
composed of large cells of uniform surface. Under the presence
of nitrogen the appearance of these cells becomes modified,
parallel striae of corrosion appearing, while the dimensions of the
cells continually diminish. vSome of the cells retain their original
appearance, while others become completely granulated. Some
are partly modified and it is possible to perceive the manner in
which the passage of the degradation from one cell to another is
resisted. When the content of nitrogen approaches 0.07 to 0.08
per cent the cells become very small, scarcely one tenth of their
original size, and at the same time the cement which fills the
separating spaces between them increases in thickness. When
454 ^^^^ Iron and Steel Magazine
this structure is developed the metal has become wholly brittle.
If the nitrogen content attains 0.2 per cent the cellular structure
wholly disappears and appears only a series of dark lines, giving
a more or less characteristic pearlitic aspect.
The cellular structure appears to bear a distinct relation to
the ductility of soft iron. The larger the cells, the more ductile
the metal. The cement which forms between the cells contains
the impurities in the metal. A content of nitrogen as high as
0.07 per cent very rarely appears in commercial products, but in
very soft irons, particularly in the products of Lancashire, a very
much lower percentage of nitrogen will suffice to render the
metal hard and brittle.
The influence of nitrogen upon hard steels is also very dis-
tinct. M. Braune discusses the behavior of a steel containing
1. 1 5 per cent of carbon, when given increasing quantities of
nitrogen by heating in an atmosphere of ammonia. At first
there is a slight increase in resistance and reduction in elonga-
tion; then, suddenly, between 0.03 and 0.035 P^^ cent of nitro-
gen, the elongation disappears entirely; the metal becomes
completely brittle. For a steel containing 0.50 per cent of
carbon the critical proportion of nitrogen corresponding to the
disappearance of ductility is 0.040 to 0.045 P^^ cent, while for
a steel of 0.02 carbon this effect is produced by 0.050 to 0.060
per cent of nitrogen. In every case the sudden attainment
of brittleness corresponds to a change in the structure.
A percentage of 0.060 nitrogen is very rare in commercial
steels, but 0.030 to 0.040 per cent frequently appears. For
this reason hard steels become brittle much more easily than
softer grades, since the proportion of nitrogen required to
cause the effect occurs more frequently in practice.
Nitrogen appears also to produce a considerable effect upon
tempered steels. The nitride of iron in such cases appears in
solution in the martensite, as it does in the ferrite for the an-
nealed steels. The influence of nitrogen upon the electric and
magnetic properties of steel is also distinct. In the case of soft
iron the coercitive force and the hysteresis are increased.
In commenting upon these remarkable researches of M.
Braune, M. Le Chatelier observes that their importance will
be evident to every metallurgist. The appearances noted in
the metallographic observations of M. Braune have been noticed
Metallurgical Notes and Commettts 455
before, but it has remained tor him to discover their origin.
An interesting fact is that the fixation of the nitrogen by the
iron does not occur directly by a combination with the nitrogen
of the atmosphere, and the presence of a basic slag appears
to be necessary as a reducing medium. This corresponds
closely with the process of the formation of the cyanides, and
indeed it has already been observed that blast furnaces which
produce much cyanide of potassium also produce an inferior
quality of iron. The nitrogen is acquired by the metal princi-
pally in the blast furnace and in the basic converter.
The experiments of M. Braune show that the fixation of the
nitrogen is effected more readily from ammonia than from the
cyanides. The ease with which the cyanides are transformed
into ammoniacal compounds in the presence of moisture renders
it probable that the vapor of water may be an intermediary in
the introduction of the nitrogen into the metal. This view is
confirmed by the experience of M. Guillet in the cementation of
nickel steel, it appearing that the use of a moist cementing
material is injurious. With a dry material for the cementation,
the interior of the case-hardened pieces shows no brittleness,
which is not the case with a moist substance.
The importance of these studies cannot be over-estimated,
and if, as appears probable, M. Braune has discovered the true
cause of brittleness in iron and steel, the way to prevent the com-
bination of nitrogen with the metal will doubtless be found by
practical metallurgists and manufacturers. After the way has
been blazed, every succeeding traveler broadens the path, and in
this, as in other departments of applied science, the operative
departments of an industry are prompt to avail themselves of the
discoveries which are made in the laboratory. If the cause of
brittleness has actually been discovered, its removal is only a
matter of. time, and the far-reaching consequences of the dis-
covery cannot now be limited.
Fractures in Large Steel Boiler Plates.* — Practically the
only material which has been used for the plates of marine
boilers for many years is mild steel. The question has been
* From a paper read by J. T. Milton at the summer meeting, July,
1905, of the British Institution of Naval Architects. Mr. Milton is chief
engineer surveyor to Lloyd's Register. " The Iron Age," August 24, 1905.
456 The Iron and Steel Magazine
recently raised whether the present tests apphed to structural
steel are sufficient to determine absolutely its quality. It may
be at once conceded that the present method of testing cannot
determine all the qualities of the steel. What the present system
does is to test the tensile breaking strength; sometimes, also,
but not often, its yield point; it also determines its ultimate
extension, its freedom from taking a temper and its capability
of withstanding cold bending. If the properties tested are all
found to be normal, it is assumed that all the other mechanical
properties will be equally satisfactory and that the material is of
good quality. Recent researches, however, show that the tests
usually applied may all yield good results, and yet the steel may
be unsatisfactory in its resistance to impact or in its endurance
of fatigue caused by repeated applications of a stress consider-
ably below the ordinary breaking strength.
In regard to the question of overheating and rolling at too
high a temperature, one of the large steel makers of this country
made the following experiment : One large ingot of boiler quality
was cut up. Three pieces were rolled into i-inch plate, one
being rolled at what is considered to be the proper temperature,
one at too high a temperature and one too cold. Pieces were also
rolled at normal temperatures and too cold into ^-inch and J-inch
plates. The pieces of plates thus made were in some cases over-
heated and allowed to cool and in other cases they were " nor-
malized " — that is, they were heated to bright red and allowed
to cool out (the ordinary method of annealing plates) ; in other
cases they had no heat treatment. They were then tested. The
results are very interesting and seem to confirm the opinions ex-
pressed by the experimenter, — namely, that when the steel is
initially good, heating the ingot between wider ranges of tempera-
ture than should occur in practice with even no more than
ordinary care does not have a very prejudicial effect on either
the ordinary mechanical tests or even on fatigue tests, the terms
" too hot " and " too cold " in these tests referring to such
extremes of temperature as would scarcely occur in actual work
without very gross carelessness. F'urther, neither does over-
heating the finished plates seem to injure them seriously. On
the contrary, in some cases it appears to have actually increased
their ductility. It must be stated, however, that the experi-
menter expresses the opinion that in plates where there is con-
Metallurgical Ahtcs and Comments 457
siderable segregation the segregated parts might behave very
differently under the various heat conditions. Some segregation
must exist in all ingots, and therefore also in all plates rolled out
of a whole ingot; but when the segregation is slight, seeing that
it must occur in the middle of the thickness of the plate near the
neutral axis as regards bending stresses, the plates, although
inferior to those without segregation, might not be unfit for use.
In view of the very large size of boiler shell plates as now
made, it is imiportant to know whether these large plates can be
made free from initial strains or whether it is inevitable that they
should have some strains in them when they are made. Seeing
that large plates can be made perfectly fiat and free from internal
strain it is reasonable that engineers should refuse to receive
those that are rolled, buckled or wavy, and should insist that in
such cases the steel makers should flatten the plates by taking
out the buckles or waves and afterward anneal the plates. Plates
should always be delivered to the boiler makers in such a condi-
tion that thev can use them with confidence without any pre-
liminary straightening treatment. Besides the bad rolling
referred to, another cause of initial stresses in plates may be
their unequal cooling on the mill floor. That this can occur is
generally considered to be improbable, but it must be admitted
that it is not impossible.
It is earnestly desired that steel makers especialty, who
have such exceptional opportunities for studying all the proper-
ties of the material they make in such large quantities, will
absolutely solve the problem why in very rare cases some
material of good sound chemical quality, made apparently in
the proper way, becomes possessed of such abnormal properties
as to become utterly unfit for the purpose for which it is made.
High-Duty Cold Saw with Teeth of High-Speed Steel.* — A
special design of high-duty circular saw which enables high-speed
steel to be used for the teeth has been designed for cutting steel
castings and forgings, structural steel, armor plate, etc., and is
illustrated herewith. The blade is a divSkof high-carbon crucible
steel, and is of sufficient stiffness to withstand any pressure
that can be safely put upon the teeth. In each face there are
slots for the teeth, and the slots are inclined backward from
* " Engineering News," October 19, 1905.
458 The Iron and Steel Magazine.
radial lines at such an angle that the pressure on the teeth when
cutting has a locking effect ; as the resultant of the pressure tends
to force the teeth inward, and as there is no tendency to draw
them by the cut, the teeth do not require to be pinned or bolted
or otherwise fastened in place. The teeth are of simple design,
forged from the bar stock in a plain die. As no machine work
is required to shape or fit them, the teeth can be made of the
modern high-speed steels, while in saws where the teeth have to
be drilled, shaped, etc., ordinary tool steel has to be used and
tempered. It will be seen also that this saw has the special
feature of dividing the cut, the teeth being placed (in staggered
position) on each side of the blade and slightly overlapped, so
that each tooth alternately takes half the width of the cut. The
teeth have a slight outward inclination, so as to make the cut
clear the saw blade, the points of the teeth being the only part
of the saw which touch the work. A broken tooth can be
quickly driven out by a chisel inserted in the semicircular recess
at the bottom of the slot, and a new tooth at once put in its place.
The Midvale Steel Company has a saw of 73 inches dia-
meter for an armor-plate cold-sawing machine. The blade or
disk is 70 inches diameter and i J inches thick, with 30 teeth on
each side, the teeth being inclined 20 degrees back of the radial
line, and having an outward inclination of i degree, so as to
make a cut i/^ inches wide. The teeth are of a special grade
of steel made by the Midvale Steel Company, and have an
angle of 60 degrees for their cutting edges. A 5 -inch nickel-
steel plate (unhardened) has been ctit with a feed of 40 inches
per hour, while a 4j-inch hardened steel armor-plate has been
cut with a feed of 9 inches per hour. The blade is bolted to a
collar on a mandrel, and the machine is directly geared to a
50-horse-power electric motor, change gears giving six peripheral
speeds of 10 to 40 feet per minute.
These saws are made by the Tindel-Morris Company, of
Eddystone, Pa. At the company's works they have been in use
for over three years without any cost for repairs. AVith machines
of sufficient power for the work, feeds of -J inch to ij inches per
minute are used on castings, forgings, bridge pins, structural
steel and other material in which the carbon does not exceed
0.40 per cent. The peripheral speed for the same kind of
material is about 50 feet per minute.
Metallurgical Notes and Comments 459
The Influence of Carbon, Phosphorus, Manganese and Sul-
phur on the Tensile Strength of Open-Hearth Steel.* — The for-
mulas established by Mr. Campbell require the use of tables in
order to take into account the influence of manganese on the
tensile strength of steel. On examining these tables I find that
the quantities given therein may be expressed by simple algebraic
forms, so that formulas for tensile strength may be written
which can be used without the help of tables.
Referring to the case where the carbon is determined by
combustion, Mr. Campbell's formula for the tensile strength of
acid steel in pounds per square inch is
40,000 + 1,000 C + 1,000 P + :\:Mn,
in which C and P are the amounts of carbon and phosphorus
in units of o.oi per cent, while rv'Mn is to be taken from his
Table VII. This table is one of double entry, the arguments
being the amounts of carbon and manganese. Taking Mn also
as the amount of manganese in units of 0.0 1 per cent, I find that
the values of .rMn given in this table are expressed by
xMn = — 320 C + 8 CMn,
and accordingly the formula for tensile strength of acid steel
becomes
40,000 + 680 C + 1,000 P + 8 CMn,
which may be used without referring to a table. For example,
let carbon be 0.50 per cent or C = 50, phosphorus be 0.05 per
cent or P = 5 and manganese be 0.45 per cent or Mn = 45 ;
then the tensile strength of the acid steel is 97,000 pounds per
square inch.
For basic steel, carbon being determined by combustion,
Mr. Campbell's formula for tensile strength is
41,500 + 770 C + 1,000 P + yM.n,
in which ;vMn is taken from his Table XIII, according to the
proportions of carbon and manganese present. For this case
I find
:vMn = — 2,700 — 120 C + 90 Mn + 4 CMn,
and hence the formula for tensile strength of basic steel be-
comes
38,800 + 650 C + 1,000 P + 90 Mn + 4 CMn,
which may be used for direct computations. For example, let
* Mansfield Merriman. "A Discussion of the paper by H. H. Camp-
bell." Transactions, American Institute of Mining Engineers.
460 The Iron and Steel Magazine
carbon be 0.30 per cent or C = 30, phosphorus be o.oi per cent
or P = I aad manganese be 0.45 per cent or Mn = 45 ; then the
tensile strength of the basic steel is 68,750 pounds per square inch.
The last term of each of these formulas contains the product
of C and Mn, and hence the formulas do not, perhaps, clearly
exhibit at first sight the influence of carbon upon the tensile
strength. They may, however, be written in the forms:
for acid steel, 40,000 + (680 + 8 Mn)C + 1,000 P;
for basic steel, 38,000 + (650 + 4 Mn)C + 1,000 P + 90 Mn.
It is now clearly seen that each o.oi per cent of carbon adds
to the tensile strength a number of pounds per square inch,
which is expressed by 680 + 8 Mn for acid steel and by 650 +
4 Mn for basic steel. Thus, if manganese is 0.50 per cent or Mn
= 50, then each o.oi of carbon adds 1,080 pounds per square inch
to the strength of acid steel and 850 pounds per square inch to
that of basic steel.
The formulas also show that each o.oi per cent of manga-
nese adds to the tensile strength a number of pounds per square
inch, which is expressed by 8 C for acid steel and by 90 + 4 C for
basic steel. Thus, if carbon is 0.30 per cent or C = 30, then
each O.OI per cent of manganese adds 240 pounds per square inch
to the strength of acid steel and 210 pounds per square inch to
that of basic steel.
The algebraic expression of the influence of carbon and
manganese on the strength of open-hearth steel is probably
only one of the important results which may be deduced from
the valuable paper of Mr. Campbell, for long-continued careful
records of actual facts will always deserve careful study.
The Minette District in France.* — The recent development
of the steel industry of France, followed by the appearance of
some of the leading works as sellers in the international markets,
is coincident with the rapid opening up of the Minette district
along the borders of Luxemburg and Lorraine. A recent report
by H. Cousin, published by the Comite des Forges de France,
presents some interesting figures which deal with the output of
the Department of Meurthe-et-Moselle in 1904. The iron
mines may be divided into two groups, that of the basin of
Nancy and that of the basin of Briey and Longwy. In 1904 the
* " Iron Age," October ly, 1005.
MetaUnrgical Notes and Comments 461
mines in the Nancy district produced 1,711,770 metric tons, as
compared with 1,668.533 tons in 1903. It is not expected that an
important increase will take place in the future. It is in the
Longwy section, with its outcrop mines, and in the Briey section,
with its deep mines, that a further rapid growth is looked forward
to. In 1904 the production of the Longwy-Briey basin was
3,821,437 tons, an increase of 588,306 tons over 1903. Adding
the output of the quarries, the production of iron ore for the
Department of Meurthe-et-Moselle was 5,951,274 tons, an in-
crease over 1903 of 658,931 tons. The shipments to other depart-
ments in France and to Belgium, Luxemburg and Germany
amounted to 1,043,000 tons. The average value at the mines
was 3.51 francs per ton. The total number of men employed
was 6,075, to whom wages aggregating 8,877,275 francs were paid.
In the Longwy-Briey basin miners average 6 to 7 francs per day,
but many of them earn more than 10 francs, or $2 per day.
Cutting machiner}^ is being employed, Morgan Gardner chain
machines being in use at the Maron-Val-de-Fer and the Mont-
Saint-Martin mines. They are electrically driven.
American Institute of Mining Engineers. — Under date of
September 22, 1905, the secretary of the American Institute of
Mining Engineers has issued a circular from which the following
is extracted :
/. — South Bethlehem Meeting, February , igo6. The nine-
tieth meeting of the Institute, for the reading and discussion of
papers, will be held at Lehigh University, South Bethlehem, Pa.,
beginning Wednesday, February 21, 1906. Further particulars
will be given by circular hereafter.
II. — Joint Meeting in England next year. The Council of
the Iron and Steel Institute has cordially invited the American
Institute of Mining Engineers to hold in England, during the
autumn of 1906, a joint meeting, consisting of sessions in London,
followed by excursions to the provinces. This invitation has
been accepted by our Council. It is understood that the meeting
will take place in August or September, but the precise date has
been left open for early determination after further conference.
Particulars concerning this and other features of the meeting
will be given in a later circular. This brief preliminary
announcement is issued in order that members may have as long
462
The Iron and Steel Magazine
a time as possible to make such arrangements for next year as
will permit their attendance.
Koristka's Illuminator for Opaque Objects. — This apparatus
is principally intended for the study of metals. It is screwed
to the microscope
tube, and contains a
total reflexive prism
which receives the
light from the front
and directs it by
means of the objective
on to the preparation.
The prism occupies
only half the field,
thus leaving the other
half free for vision.
An iris diaphragm placed in front of the prism serves to regulate
the light which it is to receive. By pulling out the arm which
carries the prism the latter may be removed from the optic field,
so as to leave it quite free. For use with this illuminator a lens
of 35 mm. diameter, and 72 mm. focus is recommended.
University of Wisconsin. — In the College of Engineering of
the University of Wisconsin a new course in chemical engineer-
ing has recently been established. It is the purpose of this
course thoroughly to train students in the fundamental principles
of both chemistry and engineering and to give such other sub-
jects as will, so far as possible, contribute toward a liberal educa-
tion.
1
Coal Production in 1904. — Mr. Edward W. Parker, of the
Division of Mining and Mineral Resources of the United States,
reports a production of all kinds of coal in the calendar year
1904 of 314,562,881 gross tons, of which 65,318,490 gross tons
were Pennsylvania anthracite and 249,244,391 gross tons were
bituminous. The value of the anthracite coal was $138,974,020
and that of the bituminous coal $305,842,268. The average
price per gross ton for the marketed sizes of anthracite coal in
1904 was $2.35 per gross ton and for the bituminous coal $1.10.
REVIEW OF THE IRON AND STEEL MARKET
October has not shown the excitement in finished material
markets which characterized September, but new business has, in
general, been greater than production. In the cruder materials,
on the other hand, there have been sharp advances, and much
excitement prevails. Standard Connellsville furnace coke has
advanced to $3.00 at ovens and is scarce even at that figure.
Pig iron has advanced a full dollar a ton since our last report,
but coke at $3.00 a ton is not in harmony even with the advanced
price of say $16.00 at furnace, as it is a dictum in the trade that a
proportion of about seven to one should be regarded as about
normal.
There is a serious shortage of crude steel, Bessemer billets
being higher, while open-hearth billets are scarcely to be had at
any price. There is less " surplus " steel than usual, a larger
percentage of ingot production going directly into finished lines
at the seat of steel production.
Never has the outlook been better in the steel trade. While
there is some uneasiness on account of the rapid adv-ances in coke
and the prospect of higher prices on pig iron, finished steel prices
have been held in line very well, a very conservative policy being
followed. It was expected that plates and merchant steel bars
w^ould be advanced, but the matter of a plate advance was
definitely settled in the negative at a meeting of the mills, while
it is well understood that the idea of advancing steel bars has
been given up.
The railroads have been heavier buyers in the past few
months than ever before, placing very large orders for rails,
locomotives and steel cars. Railroad buying is the basis of the
present extreme activity, but other consumptive lines furnish
good support. Pig-iron production in 1905 will exceed 22,000,-
000 tons, or 4,000,000 tons beyond any previous record. It is
now at substantially the maximum rate with the present equip-
ment, and it would require considerable further advance to bring
in some capacity which can be operated only under high prices.
463
464 The Iron and Steel Magazine
Pig Iron. — The whole pig-iron market has advanced about
a dollar a ton since our last report. Buying has not been heavier,
in general, than in September, but furnaces found themselves
almost sold up for this year and with large sales for next year,
and with coke advancing rapidly felt justified in advancing their
prices. The United States Steel Corporation has bought, since
about the beginning of September, a total of 120,000 tons of
Bessemer pig iron, for September, October and November
delivery, at from $14.50, valley, on up to $16.00, which price
was paid for November delivery. The corporation is likely to
buy iron for November and December very shortly. These
purchases are significant in that the}-^ take a great deal of iron out
of the market, and show that the corporation is making steel to
maximum capacity, its own production of pig iron being nor-
mally sufficient. The tonnage is small, relative to the corpora-
tion's own production, since it makes about 30,000 tons a day.
Prices are very firm as follows: F.o.b. valley furnace, Bessemer
and basic, $16.00 to $16.50; No. 2 foundry, $16.00 to $16.50;
forge, $14.75 to $15.25. Delivered Pittsburg: Bessemer and
basic, '$16.85 to $17.35; No. 2 foundry, $16.85 to $17.35; gray
forge, $15.60 to $16.10. F.o.b. Birmingham, delivery before
April i: No. 2 foundry, $13.00; gray forge, $10.75; f^i" delivery
beyond April i, furnaces are asked from 50 cents to $1.50 more.
Delivered Philadelphia: No. 2 X foundry, $17.50 "to $17.75;
standard gray forge, $15.75 to $16.00. Delivered Chicago:
northern No. 2 foundry, $17.75 to $18.00; malleable Bessemer,
$17.75 to $18.00. Freight: Birmingham to Pittsburg, $4-35;
to Cincinnati, $2.75; to Chicago, $3.65; to Philadelphia by
water, $3.50; by all-rail, $4.00.
Steel. — There has been but very little crude steel available
in the open market, while the demand has been quite heavy.
The advance in the market has been limited by what consumers
could afford to pay, having regard to prices obtainable for their
finished product. Bessemer billets can be quoted at $26 f.o.b.
Pittsburg, or $1.00 advance since our last report. Open-hearth
billets can scarcely be quoted, as there are none offered. Forging
billets would bring about $30. Sheet bars are nominally $27
for long lengths. Rods are $32 for wire rods and %:^2i for chain
rods. All prices are f.o.b. Pittsburg.
Rails. — There has been further booking of rail orders, and
Review of the Iron and Steel Market 465
with tonnage which will be held over the rail mills have about
2,000.000 tons sold for next year, while production will likely
exceed 3,000,000 tons, about the record so far, by a wide margin.
The Chicago mill is booked through November, and the Alabama
and Colorado mills about through the year, the eastern mills
being less fully sold up. The price remains at $28, f.o.b. mill,
in 500-ton lots and over. Light rails have been advancing, and
are now very firm at $26 to $27 for sections 25 to 45 pounds per
yard, the higher figure being for early delivery.
Shapes. — The market has been fairly active, a large tonnage
being booked for next year for definite construction undertak-
ings. The large mills are filled with specifications into next year,
and premiums are freely paid for small lots for early shipment.
Prices remain based on 1.70 cents for beams and channels, 1 5 -inch
and under, angles and zees.
Plates. — The Penns3"lvania Railroad system in October
placed orders for 21,000 steel freight cars, following orders for
16,000 in September, and making 37,000 cars for this system for
1906 delivery. Other roads have been good buyers, and the
steel car plants are filled until some time in the third quarter of
next year. The leading maker, the Pressed Steel Car Company,
is increasing its capacity, and next year will have a nominal
capacity of 150 cars daily. Its tonnage arrangements with the
Carnegie Steel Company are being rearranged, and beginning
January i it will receive from this company about 65,000 tons
monthly of rolled steel materials, chiefly plates, against about
45,000 tons at present. The plate mills did not advance prices
at their October meeting, and no advance is now likely for the
present, and we quote regular mill prices on the basis of 1.60
cents for tank quality, with premiums ruling for early delivery.
Merchant Bars. — The expected advance in price of steel
bars has not been made and is not likely to be considered for the
present. We quote Bessemer and open-hearth at 1.50 cents,
base, Pittsburg. Common iron bars, f.o.b. Youngstown, remain
at 1.70 cents, or 1.75 cents, Pittsburg. The Chicago market has
advanced sharply, and is now 1.80 cents, Chicago.
Sheets. — The market is now quite steady, but has not
regularly advanced. Mills are fairly well sold for a few weeks
ahead, and in some cases to the end of the year. There are still
some sellers at former quotations, 2.25 to 2.30 cents for black and
466 The Iron and Steel Magazine
3.30 to 3.35 cents for galvanized, No. 28 gauge, in carload lots
with desirable specifications, the largest lots being at a concession
of not more than 5 cents a hundred from these quotations.
Wire Products. — The market is very firm, and production is
up to the capacity of the mills. Prices have been advanced to
$1.00 a ton all around, and we now quote, in carload and larger
lots to jobbers: plain wire, 1.65 cents base; wire nails, $1.80
base.
Scrap. — Light sales have advanced heavy melting stock to
$17, delivered Pittsburg, and a still higher market is expected by
dealers who have accumulated scrap against the winter. We
quote other grades advanced as follows, delivered Pittsburg:
Cast borings, $10.25 ^^ $io-75; sheet scrap, $14.50 to $15.00;
No. I cast scrap, $15.50 to $16.00.
STATISTICS
The Production of Steel-Hardening Metals in 1904.* — Wash-
iXGTOX, D. C, September 12, 1905. — The United States Geo-
logical Survey has completed an unusually elaborate report upon
the production of steel-hardening metals in 1904, compiled by
J. H. Pratt, which shows a total output of metal ores, or con-
centrates, amounting to 945 net tons, valued at $259,620, in-
cluding the production of titanium valued at $7,000, the quantity
of which is not stated. This was a decrease in both quantity
and value as compared with 1903, but a very large increase over
1902. Over three fourths of the output of 1904 was credited
to tungsten ores. The states producing these steel-hardening
metal ores, in the order of the value of their production, together
with the metallic ore mined, are Colorado (tungsten, uranium
and vanadium), Arizona (tungsten and molybdenum), California
(chromium), Washington (molybdenum), Missouri (nickel and
cobalt) and Virginia (titanium).
In the following table is given the production in the United
States of ores of these metals for the years 1903 and 1904:
Mineral Net Tons Value Net Tons Value
Chromium 168 $2,250 138.0 $1,845
Molybdenum 795 60,865 14.5 2,175
Nickel and cobalt 661 273,900 23.0 54,000
Titanium ... ... 7,000
Tungsten 292 43,639 740.0 184,000
Uranium and vanadium. . . 30 5,625 44.5 10,600
Totals 1,946 $386,279 960.0 $259,620
Nickel and Cobalt
The main supply of nickel and cobalt produced in the United
States in the last few years has been obtained from Mine Lamotte,
Mo., as a by-product in lead smelting. During 1904, however,
* " The Iron Age," September 14, 1905
467
468 The Iron and Steel Magazine
there was no actual production of any metallic nickel or cobalt
oxide, but there was obtained 3,600 net tons of low-grade mate-
rial, valued at $54,000, ready to be smelted and refined. This
contained approximately 24,000 pounds of metallic nickel,
valued at $11,400, and 22,000 pounds of cobalt oxide, valued at
$42,600. In Virginia and North Carolina a considerable tonnage
of low-grade ore was produced in development work at deposits
located at Hemlock, Floyd County, Va., and near Webster,
Jackson County, N. C. None of this, however, was shipped
during the year.
Nearly all of the nickel used in the United States is obtained
from Canada, with a small quantity from New Caledonia. For
this reason the production of nickel ore in Canada is of especial
interest to the users of the metal in the United States. In 1903
the nickel output aggregated 12,505,510 pounds. In 1904, how-
ever, there was a falling off of approximately 2,000,000 pounds
as compared with the nickel content of the matte in 1903. A
better quality of matte, containing a much larger percentage
of nickel, has been obtained during the last few years.
Imports and Exports of Nickel
There was quite a falling off in the importation of nickel
compounds and matte, etc., during 1904, the quantity imported
into the United States in 1904 being 19,739,315 pounds, valued
at $1,122,491. As compared with the importation of 1903 of
36,217,985 pounds, valued at $1,493,889, it is a decrease of 16,-
478,670 x^ounds in quantity, but of only $371,398 in value. This
ver}^ large decrease in quantity, with only a comparatively small
decrease in value, is due to the higher grade matte that is im-
ported. The importation of cobalt oxide in 1904 amounted to
42,354 pounds, valued at $86,925. Besides this cobalt oxide
there was imported 330,983 pounds of cobalt ore and metallic
cobalt, valued at $18,272, making the total value of the importa-
tion $105,197.
As would naturally V^e expected, considering that a very
large part of the Canadian production of nickel matte is con-
sumed in this country, there is exported each year from the
United States a considerable quantity of nickel, and in 1904 this
amounted to 7,519,206 pounds, valued at $2,130,933.
Statistics 46^
Chromium
California is still the only state producing any chromite, and
in 1904 the quantity was 123 gross tons of ore, valued at $1,845.
As compared with the production in 1903 this is a decrease of 27
tons. Practically all of the chromite consumed in the United
States is imported, the greater quantity being obtained from
Turkey, with smaller quantities from New Caledonia and
Canada. The Canadian deposits of chromite are located in the
province of Quebec and in Newfoundland, and in 1904 the total
production of Canadian chrome ore amounted to 6,074 net tons,
valued at $67,146, an increase of 2,691 tons as compared with
1903. Nearly all of the chromite produced was shipped to the
United States.
Tungsten
In 1904 there was produced 740 net tons of tungsten con-
centrates, valued at $184,000, as compared with 292 tons in 1903.
The 1904 production was obtained from 10,975 "^^^s of crude ore.
During the last few years there have been small quantities of
tungsten ores and alloys imported into the United States, but as
the tungsten ores are admitted free of duty it has been impossible
to obtain the statistics for them. In 1904 the imports of a ferro-
tungsten-chromium alloy amounted to $29,439 in value, as com-
pared with $18,136 in 1903 and with $7,046 in 1902.
There has been an increasing demand for this metal during
the last year or two, stimulating the prospecting for tungsten
ores, which has resulted in the discovery of a number of new
localities where these ores are to be found. Thus tungsten,
which was formerly considered one of the rather rare elements,
has been proved to occur in large quantity and to be rather wide-
spread in its occurrence. The principal deposits found are in
Arizona, Nevada and Colorado, with others worked to but a
small extent in Idaho, Montana, New Mexico, Oregon, South
Dakota and Washington. In the Eastern states the principal
deposits are located in Connecticut, and a very small quantity
has been found in North Carolina.
The only form in which the metallic tungsten has been pre-
pared for market is as the black powder obtained by the chemical
reduction of the ores, and in this country this reduction is
carried on principally by the Primos Chemical Company, Primos,.
47 o The Iron and Steel Magazine
Pa. Fused metallic tungsten has not as yet been made com-
mercially by any of the processes by which many of these steel-
hardening metals have been obtained. Hence at the present
time tungsten steel is manufactured either by the introduction of
this powdered metallic tungsten or by the addition of the ferro
alloy. The ferro alloy cannot be produced in the blast furnace
•on account of the high temperature required, and therefore the
electric furnace has been used, in which the tungsten concen-
trates are reduced directly to the ferro alloy. Besides the ferro -
tungsten alloys there are a number of alloys of tungsten with
iron and nickel, with iron and chromium and with nickel.
Molybdenum
The production of molybdenum ores in the United States
is still very small, and in 1904 there w^as reported only 14^ net
tons, valued at $2,175. There is still considerable uncertainty
among producers of molybdenum ores as to the value of these
ores, and prices are quoted as ranging from $100 to $3,000 per
short ton. The actual value, however, of molybdenum concen-
trates at New York is probably about $200 to $250 per net ton.
The sources of supply of molybdenum are the two minerals —
molybdenite, the molybdenum sulphide (M0S2), and wulfenite,
the lead molybdate (PbMo04), In the United States the two
principal deposits of this mineral that have been thus far ex-
ploited are in the vicinity of Crown Point, Chelan County, Wash.,
and at Cooper, Washington County, Me,
Uranium and Vanadium
Although considerable development work was done in 1903
and 1904 upon uranium and vanadium deposits, the actual pro-
duction of ores of these metals was very small, amounting in
1904 to 44 J net tons of concentrates and partially concentrated
ore, valued at $10,600. The imports of uranium salts in 1904
were valued at only $9,024, as compared with imports valued at
$13,498 in 1903 and at $12,491 in 1902.
Of these two metals, uranium and vanadium, it is only the
latter that has been used in the manufacture of steel, although
uranium has been tested and experimented with to some slight
extent for this purpose. No large quantity of vanadium steel
i
Statistics 471
has been made and there is but very Httle of it on the market,
although the ferro alloys are now being made and offered for sale.
The metal uranium is included with the steel-hardening metals on
account of the experiments that have been made with it for this
purpose; also on account of its close relation to vanadium and
because so many of the ores of vanadium contain uranium. All
of the uranium minerals have become of special interest in the
last year or two since the discovery of radium, as uranium seems
to be the source of radium.
Titanium
The production of titanium, or rutile, during the year 1904
was valued at approximately $7,000. Although the actual
commercial value of titanium as a steel and iron hardening metal
has not as yet been thoroughly demonstrated, still the experi-
mental work that has been done seems to indicate that titanium
will become of some importance in the production of both coke
and charcoal iron.
The World's Production of Finished Iron.* — In our issue of
July 20 reference was made to the production of rolled iron in the
United States in 1904, which amounted to 1,760,084 gross tons.
This total compared with 2,518,194 tons for 1890, the last pre-
ceding year in which the statistics of rolled iron production had
been gathered separately from those of steel. The London
" Iron and Coal Trades Review " combines these figures with
those of other countries and to the discomfiture of the prophets of
the extinction of the puddling industry. The production of
rolled iron in leading countries in the past two years is thus
stated :
1903 1904
United Kingdom (gross tons) 950,390 936,228
Germany (metric tons) 819,832 765,197
France (metric tons) 589,910 554,632
Belgium (metric tons) 401,550 360,520
Russia and other European countries (metric
tons) 850,000 800.000
Totals 3,61 1,682 3,416,577
Add United vStates (gross tons) 1,760,084
Grand total 5,176,661
* " The Iron Age," August 24, 1905.
47 2 The Iron and Steel Magazine
Estimating that with ah countries included the world's
production of rolled iron was about 6,000,000 tons last year, our
contemporary points out that there has certainly been no ex-
tinction of the finished iron industry and that these products are
evidently as much appreciated as they ever were and deemed
quite as indispensable.
In 1880 the output of finished iron in the United Kingdom,
the United States, Germany, France, Belgium, Austria-Hungary,
Russia and Sweden was 8,553,225 tons, and in 1890 8,340,599
tons. Last year the world's output of steel ingots was about
36,000,000 tons, so that the ratio of steel to iron was substantially
6 to I.
RECENT PUBLICATIONS
The Crystallization of Iron and Steel, by J. W. Mellor. 144 4 X
7J-in. pages; illustrated. Longmans, Green & Co. New York.
1905. Price , $1.60. — In this little book the author has attempted
with a marked degree of success, to present clearly and concisely
the fundaments of modem metallography. The subject is treated
in a logical order, the author describing in turn the constitution
of alloys in general, the constitution of iron and steel, the rational
of the hardening, annealing and tempering of steel, and the in-
fluence of stress and strain. This is followed by a description of
the preparation of samples of metals for microscopical examina-
tion and by the glossary of terms drawn in 1901 by the Iron and
Steel Institute. The book also includes a well prepared index.
The photomicrographs reproduced to illustrate typical struc-
tures have generally been selected with care and are chiefly from
Arnold, Stead and Osmond. On page 15 and others, the author,
in explaining the allotropic theory, evident!}^ attributes the
hardness of suddenly cooled steel to the retention of gamma iron,
whereas in the light of Osmond's recent work, gamma iron (aus-
tenite) is relatively soft, the hardness of quenched steel being
due to the presence of beta iron in the form of martensite. The
author in a foot-note, however, expresses his doubt as to whether
it is gamma or beta iron which renders steel hard, but he adds
that these are both supposed to be hard. On page 54 he states
that " gamma iron is said to be as hard as chilled steel." The
illustration facing page 23 and representing the structure of
granular pearlite (sorbite) after Heyn, was not selected with
the care shown in the choice of the other photomicrographs,
seeing that many photomicrographs of this kind have been
published of much greater excellence. It is, we think, to be
regretted that the author has not used the happy term " eutec-
toid " in place of " eutectic " and " saturated " to designate
steel having a eutectic-like constitution that is made up exclu-
sively of pearlite, as suggested by Professor Howe. On page 32
473
474 The Iron and Steel Magazine
the author gives a classification of substances which should be
very helpful to students of metallography as it brings out clearly
the nature both of eutectic mixtures and of solid solutions. The
classification is as follows:
Heterogeneous \ Indefinite proportions Ordinary mixture
} Definite proportions Eutectic mixture
I One component Element
Homogeneous J I Definite proportions Chem. compounds
( More than one \ -r , r- -^ , . S Liquid Liquid solution
( Indefinite proportions ^ Solid Solid solution
On page t^t, and others the solid solution of carbon in iron (in
gamma iron) which is stable at a high temperature (above the
critical range) is called martensite by the author, whereas we
believe that this constituent should be called " austenite "
according to the views of Osmond and other authoritative
writers, and the name of martensite reserved for the solid solu-
tion of carbon in beta iron, the ordinary constituent of hardened
steels. It is much to be regretted that such confusion and
divergence of opinion should still exist among metallographists
on so vital a point, while there seems to be no good reason for
it. On page 64 Arnold's experiments by which he endeavored
to show that there was no marked increase in the tenacity at
the Ar3 and Ar2 points are apparently indorsed by the author,
who ignores the discussion of these experiments by which it was
revealed that on the contrary a decided and abrupt increase of
tenacity did take place at these points. On page 67, referring
to a steel crystal from Tschemoff's collection, 15 inches is given
as the equivalent of 15 cm., evidently through a typographical
error. The length of the crystal was about 15 inches. On
page 69, it is stated that when an hypereutectoid steel cools
*' the pearlite behaves like a pure metal, and rejects the excess
of cementite to the boundaries, so as to form a network of
cementite. Is it not cementite that forms first (falls out of
solution), and would it not be more correct to say that the excess
of cementite is rejected by the austenite (the mother metal) and
that later each mesh of austenite (which has now reached the
nature of hardenite) changes in situ into pearlite?
This book is a very valuable addition to metallographic
literature and it may be warmly recommended as a textbook
for all students interested in this growing subject.
Metallurgy of Cast Iron, by Thomas D. West. Ninth edi-
tion. 677 5 X 7 -in. pages; illustrated. The Cleveland Printing
Rccctit Publications 475
and Publishing Company. Cleveland, Ohio. 1904. Price, $3.00.
— The fact that this is the ninth edition of this well-known
book is in itself a testimony of its worth and popularity.
Lalwratory Chemistry, by Richard B. Moore. 195 5 X 7-in.
pages; illustrated. J. B. Lippincott & Co. Philadelphia. 1904.
Price, 75 cents. — The purpose of this little book is to briefly
describe the fundamental principles of physics and chemistry
and to illustrate them by means of simple experiments to be
performed by the student. The book is written for the use of
students in secondary schools.
Laboratory Notes on Practical Metallurgy, by Walter Mac-
Farlane. 140 5 X 7-in. pages; illustrated. Longmans, Green &
Co. New York. 1905. — In this excellent little book the author
describes one hundred and twent3^-seven experiments illustrat-
ing chemical principles of interest to metallurgists. They are
for the most part of a simple character and such as can readily
be performed with the ordinary equipment of the chemical
laboratories of technical schools.
Coke, a treatise on the Manufacture of Coke and Other
Prepared Fuels and the Saving of By-products, by John Fulton.
476 6 X 9-in. pages; illustrated. International Textbook Com-
pany. Scranton, Pa. 1905. Price, $5.00. — In this, the second
and revised edition of his book, the author has brought up to
date the description of an industry of vast and growing magni-
tude. The subject is divided- into eleven chapters which are
fully and carefully illustrated, while the typography of the book
is very satisfactory. That this is the best and most exhaustive
treatise, in the English language at least, dealing with the
manufacture of coke will, we believe, be generally conceded.
The Copper Handbook, by Horace J. Stevens, Volume V
(1904). 882 6 X 8r5-in. pages; illustrated. Houghton, Mich.
1905. Price, in buckram binding, $5.00; in full morocco, $7.50.
— Each annual issue of this valuable puVjlication is more exhaust-
ive than the preceding one. The present volume is divided into
sixteen chapters, dealing respectively with the history, geology,
chemistr^^ mineralogy, metallurgy and uses of copper, glossary
47 6 The Iron and Steel Magazine
■of mining terms, copper deposits of the United States, of Canada
and Newfoundland, of Europe, of Africa, of Asia, of Australia
and Oceanica, copper mines of the world and statistics of copper.
This is followed by a carefully prepared index.
In the present edition the two final chapters which are the
most important and constitute five fifths of the entire book have
been completely rewritten.
The fifteenth chapter contains no less than 3,849 titles,
with from two lines to fourteen pages devoted to each. The
editor writes that of the many companies denounced as dis-
honest or downright fraudulent in the last edition of this book,
not one has been able to prove its right to a better rating than
was accorded it, though a number have made attempts to secure
such ratings, by methods ranging from covert bribery, through
legal proceedings, down to threats of physical violence. Of the
many companies indorsed as honorable in the past four annual
editions of the copper handbook, only one has been found dis-
honest. It will be apparent that the contents of this book are
invaluable to all those interested in copper, whether their interest
be technical or financial.
Friction and Lubrication, by William M. Davis, second edi-
tion. The Lubrication Publishing Company. Pittsburg, Pa. —
The important subject of lubrication and lubricants is treated
in this book with much authority and in a thoroughly practical
manner, the book being written essentially for the mechanical
man. The author tells us that he has kept in mind the fact that
engineers and managers and mechanics are busy men and that
he has therefore tried to present the matter in a- plain, concise
way, that will be readily understood by readers and be of practical
value in their every-day work.
Machine -Shop Tools and Methods, by W. S. Leonard. 554
6 X 9-in. pages; nearly 700 illustrations. John Wiley & Sons.
New York. 1905. Price, $4.00. — This book was written prin-
cipally as a textbook for his students by the author, who is
instructor in machine-shop practice in the Michigan Agricultural
College, but it will be found of great value by all those interested
in machine-shop work. It undoubtedly fills a place in techno-
logical literature not heretofore occupied. Tools, machines and
Recent Publications 477
manipulations are clearly and methodically described, while the
book is finely printed, illustrated and bound. Three hundred
and forty-seven well selected questions are appended to the
book and should prove of much value to the student.
Mechanics of Materials, by Mansfield Merriman. Tenth edi-
tion, rewritten and enlarged. 507 6 X 9-in. pages; illustrated.
John Wiley & Sons. New York. 1905. Price, $5.00. — This
is the tenth edition of Professor Merriman's well-known book.
In the present edition the book has been rewritten and enlarged
from 329 to 518 pages. According to the publisher's notice, the
main purpose in rewriting the book has been to keep it abreast
w4th modem progress, but the attempt has also been made to
present the subject more clearly than before, in order to advance
the interests of sound engineering education and to promote
sound engineering practice. This new edition of so important
and popular a work will undoubtedly be welcomed by the engi-
neering profession. The name of the publisher is a guaranty of
excellent typography, binding and general make-up of the book.
BOOKS RECEIVED
The following books have been received and will be reviewed in an
early issue of the Iron and Steel Magazine.
Cours d' Exploitation des Mines, third edition, by Haton de la Goupil-
liere, with revisions and addition by Jean Bes de Berg. Volume I. 1002
6\ X lo-in. pages; illustrated. Paper covers. Vve. Dunod. Paris.
1905. Complete in three volumes. Price, 90 francs.
Technological and Scientific Dictionary. Part X. Edited by G. F.
Goodchild and C. F. Tweney. 64 7 X lo-in. pages. Paper covers. George
Xewnes. London. Price, one shilling. The work will be complete in
15 parts.
The Universal Directory of Railway Officials, 190 jy. Compiled under
the direction of S. Richardson Blundstone, editor of the " Railway
Engineer." 667 5I X 8^-in. pages. The Directory Publishing Company.
London. Price, 10 shillings.
Transactions of the Institution of Mining and Metallurgy. Volume
XIII (1903-1904). Edited by Arthur C. Claudet and C. McDermid.
568 6x8i-in. pages; illustrated. Paper covers. E. and F. N. Spon.
London.
Der Eisenbau, Luigi Vainello. 691 6 X 8i-in. pages; illustrated.
R. Oldenbourg. Munich and Berlin. 1905. Price, 17.50 marks.
478 The Iron and Steel Magazine
Technical Methods of Ore Analysis, by Albert H. Low, 273 6 X 9-in.
pages; illustrated. John Wiley & Sons. New York. 1905. Price ^
$3.00.
Contribution a V Etude de la Fragilite dans les Fers et les Aciers (Con-
tribution to the Study of Brittleness in Iron and Steel). 482 9 X n-in.
pages; illustrated. Paper covers. Societe d' Encouragement pour
rindustrie Nationale. Paris. 1904. Price, 20 francs.
Smithsonian Institution. Annual Report of the Board of Regents for
the year ending June 30, 1904. 804 6 X 9-in. pages; illustrated. Govern-
ment Printing Office. Washington, D. C. 1905.
PATENTS
RELATING TO THE METALLURGY OF IRON AND STEEL
UNITED STATES
798,723. Malleableizing and Annealing Oven. — William L.
Casaday, South Bend, Ind.
798,834. Pouring Device for Ladles, Etc. — Reinhold Schnei-
der, Sharon, Pa.
799,001. Furnace for the Manufacture of Iron Sponge. —
Gustaf Groendal, Djursholm, Sweden.
799,189. Producing Wrought-Iron Sponge. — Dexter Reynolds,
Albany, N. Y.
799,269. Rolling-Mill Roll. — Ambrose Ridd, Newport, Ky.,
assignor of one half to Albert Simms, Newport, Ky.
799,542. Process of Cementing Iron or Steel. — Charles C.
Davis, Germantown, Pa.
799,634. Production of Metallic Strip, Wire, Rods, Etc. —
Sherard O. Cowper-Coles, London, England.
^799,860. Process of Galvanizing Wire. — Guy L. Meaker,
Evanston, 111., assignor to the American Steel and Wire Company of New
Jersey, a corporation of New Jersey.
799.861. Electrolytic Apparatus. — Guy L. Meaker, Evanston,
111., assignor to the American Steel and Wire Company of New Jersey, a
corporation of New Jersey.
799.862. Process of Separating Ferriferous Zinc Compounds.
— Guy L. Meaker, Evanston, 111., assignor to the American Steel and Wire
Company of New Jersey, a corporation of New Jersey.
799,916. Feed-Train for Rolling Mills. — Emil Meyer, Duis-
burg, Germany.
800,018. Annealing Furnace. — Ambro.se Ridd, Newport, Ky.
800,698. Preparing Fine Particles of Oxide of Iron for Use in
Furnaces. — Utley Wedge, Ardmore, Pa.
800,712. Ingot Stripper. — John I. Blount, Donora, Pa., assignor
to Whiting Foundry Equipment Company, Harvey, 111., a corporation of
Illinois.
800,857. Electric Furnace. — Fredrik A. Kjellin, Saltsjoebaden,
Stockholm, Sweden.
GREAT BRITAIN
19,464 of 1904. Briquetting Ore. — C. Reinke, Bredelar, Germany.
For briquetting tine iron ores, the use of a mixture of limestone containing
479
48 o The Iron and Steel Magazine
a high percentage of carbonate of lime, and Portland cement as binding;
material.
20,842 of 1904. Grate-Bar Alloy. — W. G. Crosthwaite, Leeds.
An alloy of iron and aluminum for use in making firebars in furnaces,
made by adding to 10 tons of pig iron, 2 cwts. of aluminum and 10 cwts, of
scrap steel; will not burn away so quick as ordinary cast iron.
6,388 of 1905. Briquette Furnace. — F. J. Bergendal, Herraeng,.
Sweden. An improved furnace for burning briquettes made of fine iron
ores.
14,214 of 1905. Electric Furnace. — F. A. Kjellin, Stockholm,.
Sweden. An improved method of arranging the electric fittings of electric
furnaces, so that they do not get damaged by the heat of the furnace.
21,538 of 1904. Disposition of Flue Dust. — A. Sahlin, London.
Apparatus for removing dust from blast-furnace and similar gases, con-
sisting of a revolving horizontal drum containing baffle-plates and water.
22,700 of 1904. Briquetting Machine. — J. F. Pease, Darlington.
Improvements in the mechanism of machines for making briquettes of
iron ore, of the kind in which there are a succession of molds mounted on
a circular frame and plungers to work into them.
I
SIR HENRY BESSEMER
^ ^/
The Iron and Steel Magazine
'* Je venx au mond publier
d'une phimc de fer stir nn papier d'acier.'^
Vol. X December, 1905 No. 6
THE GENESIS OF THE BESSEMER PROCESS*
TT will, perhaps, assist the non -technical reader to understand
what follows if I explain, in a few words, the forms in which
iron and steel existed at the time when I commenced the experi-
ments which resulted in the creation of the Bessemer process.
At that date there was no steel suitable for structural purposes.
Ships, bridges, railway rails, tires and axles were constructed of
wrought iron, while the use of steel was confined to cutlery, tools,
springs and the smaller parts of machinery. This steel was
manufactured by heating bars of Swedish wrought iron for a
period of some six weeks in contact with charcoal, during which
period a part of the carbon was transferred to the iron. The
bars were then broken into small pieces and melted in crucibles
holding not more than sixty pounds each. The process was
long and costly, and the maximum size of ingot which could be
produced was determined by the number of crucibles a given
works could deal with simultaneously. Such steel when rolled
into bars was sold at £50 to ^£6o a ton. The wrought -iron bars
from which the steel was made were manufactured from pig iron,
as was all wrought iron, by the process known as " puddling."
Naturally, such a process was costly; puddling demands great
strength and endurance on the part of the workmen, combined
with much skill.
Practically, all objects in iron, except such as were simply
castings, were at that time made from wrought iron manufac-
* From Sir Henry Bessemer's recently published autobiography,
through the courtesy of the publishers, " Engineering," London.
482 The Iron and Steel Magazine
tured by puddling. The object I set before myself was to pro-
duce a metal having characteristics comparable with those of
wrought iron or steel, and yet capable of being run into a mold
or ingot in a fluid condition. I was aware that Fairbairn and
others had sought to improve cast iron by the fusion of some
malleable scrap, along with the pig iron, in the cupola furnace.
This fusion of scrap iron, intermixed with a mass of coke, was
found to convert the malleable iron into white cast iron, which
was at the same time much contaminated with sulphur. There-
fore, to a great extent, this s^^stem had failed in its object. In
my experiments I avoided the difficulties inseparable from
Fairbairn 's method, by employing a reverberatory furnace in
which the pig iron was fused. Into the bath so formed I put
bro ken-up bars of blister steel, made from Swedish or other
charcoal-iron, its fusion taking place without its being further
carburized by contact with the solid fuel or contaminated by
the absorption of sulphur. The high temperature necessary for
the fusion of a large proportion of steel in the bath was obtained
by constructing the fire grate much wider than the bath, by
contracting the width of the furnace considerably at the bridge
and also by continuing to taper slightly the furnace all the way
from the fore bridge to the downcast flue, which was connected
with a tall chimney shaft. Many alterations and modifications
of this furnace were made from time to time, but it was found
that the large volume of flame sweeping over the open hearth of
the furnace was mixed with a considerable quantity of com-
bustible gas. To consume this gas a hollow fire bridge was
employed, having numerous perforations made in the clay
lumps of which it was composed, and so arranged as to allow
jets of hot atmospheric air to mingle with these combustible
gases, and produce an intense heat close down to the surface
of the bath. It was also found that this admission of hot
air all along the back of the fire bridge produced a decarbonizing
action on the bath; hence, the state of carburation of the
metal might be altered by regulating the admission of air. This
passage of air through the hollow fire bridge served also to keep
down the temperature of the latter and render it more durable.
Some of the samples of metal which I produced were, when
annealed, of an extremely fine grain and of great strength. At
this stage of my experiments I cast a small model gun, which
TJic ( J cues is of the Bessemer Process 483
in tlio lathe gave shinings slightly curled, and closely resembling
the turnings from a steel ingot: the metal, when polished, also
looked white and close-grained like steel. I was so well pleased
with this little model gun that I took it over to Paris, obtained
an audience with, and showed it to the emperor, who had en-
couraged this attempt to improve the iron employed in founding
heavy ordnance. His Majesty, who had desired me to report
progress, accepted this experimental gun, remarking that some
day it might have an historical interest. It was in recognition
of this circumstance that his Majesty, later on, intimated
through Colonel Belleville, his desire to confer on me the decora-
tion of the Legion of Honor, provided I could obtain permission
to wear it, a privilege which our ambassador twice refused.
His Majesty also sanctioned the erection of my furnace at the
Government Cannon Foundry at Ruelle, near Angouleme, to
which place I went with proper introductions for the purpose of
arranging all the necessary details. I also sent over from
England several thousand special fire bricks, etc., for the erection
of the furnaces.
But, on resuming my further researches, after my return to
London, an incident occurred which suddenly put a stop to the
intended works at the Ruelle gun foundry, and in fact altered
all my future plans and investigations.
The furnace, as then arranged, is shown in vertical section
in Fig. 35, and in horizontal section, on the line passing above
the fire bridge, in Fig. 36, Plate XII, the bath being empty and
showing the tapping hole and the way in which the furnace
narrows at the fire bridge. Fig. 37, on the same plate, is also a
horizontal section, taken on a line passing through the openings
in the perforated hollow fire bridge, and clearly shows how the
jets of air were directed so as to produce an intense ignition of
the combustible gases, mingled with, and passing over with, the
large volume of flame from the overcharged fire grate.
The small scale on which this experimental furnace was
built (a capacity of 3 cwt. only) was much against my obtaining
the high temperature necessary to melt a large proportion of
steel in a pig-iron bath. I was, of course, fully aware that a
furnace of sufficient capacity to cast a 5-ton or a lo-ton gun
would acquire a much higher temperature than was possible in
my small furnace. I knew, also, that forced draft obtained by
484
The Iron and Steel Magazine
ri.AiK XII
;;:). N'Kurii ai. SKdiuN ok Kl:i!X.\('K !'.i; Makim, Mai,ij:ai;!.i-; 1k<>n
^=W"
:i(;. lii ii;l/M\ ! \l, SKCTItiN ul- i'lUNAt
,_,
.MAklNt. M \IJ,KAi;l.K li:"N
^^iiw^wmw *,- *"'"■>■
Fig. 37. Horizontal Section of Furnace for Makin.t,^ Malleable Iron
The Geiicsis of the Bessemer Process 485
closing in the ash pit and forcinL^- air into it would still further
increase the temperature. That this forced draft was in my mind
at the time is shown by the fact that I took out a patent for the
manufacture of cast steel, dated October 17, 1885 ; that is, about
two months after the casting of the model gun, in which specifica-
tion I fully described the forcing of air by a fan into the closed ash
pits of the furnaces employed in the manufacture of cast steel.
It has since often occurred to me that, with these additional re-
sources still untried, I did not act wisely in so suddenly abandon-
ing these open-hearth experiments, in favor of an entirely differ-
ent system, suggested to m}- mind by the incident to be presently
referred to. But with my impulsive nature, and intense desire
to follow up every new problem that presented itself, I at once
threw myself unreservedly into this new study, which seemed
to open the way to the rapid production of bars, rails and plates
of malleable metal direct from the blast furnace.
Before dismissing this subject, it may be interesting, even
at this distant period, to speculate on what would have been the
natural outcome of my open-hearth furnace experiments, had
I not been so suddenl}^ diverted from their further pursuit.
Such a furnace, with forced draft and a capacity of ten
tons, would undoubtedly have melted malleable iron or steel in
a bath of pig iron, and have decarburized the latter to the desired
extent; for I had, in fact, already fused steel in a bath of pig
iron on the open hearth of this small reverberatory furnace,
and as far back as January, 1855, I had claimed in my patent:
The fusion of steel in a hath of melted pig or cast iron in a rever-
beratory furnace, as herein described .''
This was about ten years prior to the patent taken out by
M. Emile Martin, and now generally known as the " Siemens-
Martin process." This latter patent was obtained in England
in the name of A. Brooman, the patent agent of Emile Martin,
and is dated August 18, 1865, or more than ten years after my
patent of January 10, 1855. M. Emile Martin in his patent
says: " The manufacture is effected upon the principle of fusion
of iron or natural steel in a bath of cast iron, maintained at a white
heat in a reverberatory furnace, such as Siemens gas furnace."
I, however, desire to say that I make no claim to the prior
invention of the Siemens-Martin process, nor do I assume that
my patent of 1855 furnished any information which either of
486 The Iron and Steel Magazine
these gentlemen had availed themselves of, although my patent
for melting steel in a bath of cast iron on the hearth of a rever-
beratory furnace had been granted, and the specification pub-
lished, some nine years prior to M. Martin's application for his
patent. But seeing how many years I was in advance of M.
Martin, I feel perfectly justified in saying that the fusion of steel
in a bath of pig iron on the open hearth of a reverberatory fur-
nace, which I had patented and accomplished ten years prior
to the Siemens-Martin patent, was, to use a favorite expression
of Mr. Gladstone, " approaching within measurable distance "
of that successful process known as the open-hearth manufacture
of mild steel.
On my return from the Ruelle gun foundry I resumed my
experiments with the open-hearth furnace when the remarkable
incident mentioned above occurred in this way. Some pieces
of pig iron on one side of the bath attracted my attention by
remaining unmelted in the great heat of the furnace, and I turned
on a little more air through the fire bridge with the intention of
increasing the combustion. On again opening the furnace door,
after an interval of half an hour, these two pieces of pig still
remained unfused. I then took an iron bar, with the intention
of pushing them into the bath, when I discovered that they were
merely thin shells of decarburized iron, as represented at A,
Fig. 37, Plate XII, showing that atmospheric air alone was cap-
able of wholly decarburizing gray pig iron, and converting it
into malleable iron without puddling or any other manipulation.
Thus a new direction was given to my thoughts, and after due
deliberation I became convinced that if air could be brought
into contact with a sufficiently extensive surface of molten crude
iron, it would rapidly convert it into malleable iron. This, like
all new problems, had a special interest for me, and I became
impatient to test it by a laborator}^ experiment. Without loss
of time I had some fire-clay crucibles made with dome-shaped
perforated covers, and also with some fire-clay blow pipes, which I
joined on to a 3-foot length of i-inch gas pipe, the opposite end
of which was attached by a piece of rubber tubing to a fixed
blast pipe. This elastic connection permitted of the blow pipe
being easily introduced into and withdrawn from the crucible,
as shown at Fig. 38, Plate XIII, which represents a vertical
section of an air furnace containing a crucible that, in this case,
TJic Genesis of iJic Bessemer Process 487
fomis the " converter." About ten pounds of molten gray pig
iron half filled the crucible, and thirty minutes' blowing was
found to convert 10 ])ounds of gray pig into soft malleable iron.
Here at least one great fact was demonstrated, viz., the absolute
decarburization of molten crude iron without any manipulation,
hut not without fuel, for had not a very high temperature been
kept up in the air furnace all the time this quiet blowing for
thirty minutes was going on, it would have resulted in the solidi-
fication of the metal in the crucible long before complete decar-
burization had been effected. Hence arose the all-important
question: Can sufficient internal heat be produced by the intro-
duction of atmospheric air to retain the fluidity of the metal
until it is wholly decarburized in a vessel not externally heated?
This I determined to try without delay, and I fitted up a larger
blast cylinder in connection with a 20 horse-power engine which
I had daily at work. I also erected an ordinary founder's cupola,
capable of melting half a ton of pig iron. Then came the
question of the best form and size for the experimental " con-
verter. ' ' I had very little data to guide me in this, as the crucible
converter was hidden from view in the furnace during the blow.
I found, however, that slag was produced during the process
and escaped through the holes to the lid. Owing to this, I
determined on constructing a very simple form of cylindrical
converter, about four feet in height in the interior, which was
sufficiently tall and capacious, as I believed, to prevent any-
thing but a few sparks and heated gases from escaping through
a central hole made in the flat top of the vessel for that purpose,
as shown in the vertical section at Fig. 39, Plate XIII. The
converter had six horizontal tU3^eres arranged around the lower
part of it; these were connected by six adjustable branch pipes,
deriving their supply of air from an annular rectangular chamber,
extending around the converter, as shown.
All being thus arranged, and a blast of 10 or 15 pounds pres-
sure turned on, about 7 cwt. of molten pig iron was run into the
hopper provided on one side of the converter for that purpose.
All went on quietly for about ten minutes ; sparks such as are com-
monly seen when tapping a cupola, accompanied by hot gases,
ascended through the opening on the top of the converter, just
as I supposed would be the case. But soon after a rapid change
took place; in fact, the silicon had been quietly consumed, and
488
PLATE XIII
The Iron and Steel Magazine
' ji^' ;■'■■■"' '^', ■*'*' iSf'»S'
SkciI'ix "i t 'i;r( ir.i.i:
Willi r.i.MW I'li'K
ii;. ;!!*. SiciTioN 111 \'kI!I ii',\i ( 'uw Kinicj;
|ip<p>»i>iiiiii(iwn»<
VU.. II. I'l'.. H'.
1M(;«. IU AM) 11. SKCTIUNS Ol' VkHTICAI, CuNViCltTKK WITH I 1'1'KH Cu AM liKU
The Geiissis oj the Bessefuer Process 489
the ox\-!j^en, next unitiiii^' with the earbon, sent up an ever-
inereasing stream of sparks and a volummons white flame. Then
followed a succession of mild explosions, throwing molten slags
and splashes of metal high up into the air, the apparatus becom-
ing a veritable volcano in a state of active eruption. No one
could approach the converter to turn off the blast, and some
low, flat, zinc-covered roofs, close at hand were in danger of
being set on fire by the shower of red-hot matter falling on them.
All this was a revelation to me, as I had in no way anticipated
such violent results. However, in ten minutes more the eruption
had ceased, the flame died down and the process was complete.
On tapping the converter into a shallow pan or ladle, and form-
ing the metal into an ingot, it was found to be wholly decar-
burized malleable iron.
Such were the conditions under which the first charge of
pig iron was converted in a vessel neither internally nor exter-
nally heated b}^ fire.
I, however, desired to convert a second charge of pig iron
which had been put into the cupola; and in order to prevent
this dangerous projection upwards of sparks and molten slags,
a temporary expedient was resorted to, which, however, failed
in its object.
I procured one of those circular, checkered cast-iron plates
so much used in the London pavements to allow coals to be put
in the cellars below the pavement. This plate which was about
a foot in diameter, was suspended by a chain at a distance of
about 18 inches above the central opening in the top of the con-
verter, as shown in Fig. 30, Plate XIII.
This, as a mere temporary device, was deemed sufficient
to allow the conversion of another 7 cwt. charge to be effected
without any danger of setting fire to the premises. The convert-
ing operation went on quietly as before, but when the eruption
commenced, I saw the suspended plate get rapidly red hot, and
in a few minutes more it melted and fell away, leaving the chain
dangling over the opening, and allowing the slags and splashes
of metal to shoot upwards as before. Thus it happened that
the first converter that I constructed was at once condemned
as commercially impracticable, owing to this vertical eruption
of cinder, and for this reason only.
All attempts to lessen the violence of the process t)y the
490 The Iron and Steel Magazine
reduction of the number of tuyeres, or by lessening their diame-
ter, or by diminishing the pressure of the blast, only resulted
in a reduction of the necessary temperature, and in preventing
the conversion of the molten pig into malleable iron. In one
case the trial of a diminished area of tuyere openings resulted
in nearly the whole charge of metal, after more than an hour's
blowing, being converted into a solid mass of brittle white iron,
similar to ordinary refiner's plate metal. Indeed, I may say
the result of all my early investigations proved to me, beyond
the possibility of a doubt, a fact which has since been confirmed
in every Bessemer steel works throughout Europe and America,
viz., that rapidity of action, ending in a violent eruption, is an
absolutely necessary condition of success. Not only must the
converted metal acquire an enormously high temperature, so
that it may not be chilled when pouring it out of the converter,
or when a relatively large quantity of much cooler metal be
added to deoxidize it, but it must not chill and form a shell in
the ladle during the comparatively long time required for casting
the ingots. Hence, to carry out the Bessemer process success-
fully, a temperature must be obtained very considerably above
the mere melting temperature of malleable iron; and in order
to secure this it is necessary to drive powerful streams of air
into the metal, so as to divide it into innumerable tiny globules
diffused throughout the whole body of iron under treatment
which, for the time being, may be likened to a fluid sponge with
the active combustion of carbon with oxygen going on in every
one of its myriads of ever-changing cavities.
It has been found that the union of carbon and oxygen takes
place so rapidly at this high temperature as to produce a series
of mild explosions. In the large converters in common use, a
space some 8 feet or lo feet in height above the normal level of
the metal is provided, in which this violent action expends itself
unseen, and is only partially recognized by a .small quantity of
slags leaping out of the mouth of the converter.
With these facts before us, it must be self evident that all at-
tempts to produce malleable iron in a plain cylindrical vessel that
has no top to it, and in which the metal normally rises to with-
in 6 inches of the open mouth, must utterly fail from two causes:
first, because heat would fly off so freely that the temperature
of molten malleable iron could never be reached; and secondly,
Tjic Genesis of the Besse))ier Process 491
because nearly all the metal contained in such a shallow, open-
topped vessel wcHild ha\c leaped out of it, and have been scat-
tered in all directions on the occurrence of the explosive eruption,
without which no charge of molten pig iron has, or can be, con-
verted into fluid malleable iron by a blast of air.
I had no sooner condemned m}^ first cylindrical converter
than I commenced to remedy its defects. The most obvious
and ready way of doing this would have been simply to make
an opening on one side of it near the top, and thus allow the
escape of the ejected matter to take place horizontally, directing
it against a wall, or allowing it to fall into a pit. But I desired
to prevent this discharge of metal splashes as much as possible.
Hence I determined on constructing a new converter with an
upper chamber, having an arched roof and a conical sloping
floor. This converter is represented in Figs. 40 and 41, on Plate
XIII, the last-named view being a horizontal section through
the tuyeres. When a converter is so constructed, the ejected
fluid that would otherwise pass vertically upwards into the air
is thrown against the arched roof, and any metal that may be
emitted falls again on the sloping floor of the upper chamber
and returns to the lower one. The flame and a portion of the
slags find their way out of the two square lateral openings pro-
vided for that purpose. This upper chamber also served as a
receptacle for heating up any metal intended to recarburize, or
alloy with, the steel in course of being converted. The sectional
plan, Fig. 41, shows six well-burned fire-clay or plumbago tuyere
pipes fitted to openings left in the lining for that purpose. Their
outer ends were made conical to facilitate the ramming in of
loam around them, which effectually held them in position, and
at the same time admitted of their easy removal when worn out ;
a jointed piece of iron tube, with a catch to hold it in place,
conveyed the blast to each tuyere.
Another view. Fig. 42, Plate XIV, of this converter, taken
at right angles to Fig. 40, shows on one side the hopper by which
the molten iron was run into it by a movable spout direct from
the cupola. This view also shows the tapping-hole open, and the
spout which conducted the converted metal into a movable
shallow pan or receiver, supported by a long handle (not shown).
A fire-brick plug attached to a long handle was fitted to a fire-
brick ring or opening in the bottom of the pan, and prevented
The Iron and Steel Magazine
Fig. 42. Section of Converter, Ladle and TTydranlic Ingot Mold
The Ge)icsis of the Bessemer Process 493
any debris from the tapping hole boini^- carried into the mold.
As tkis apparatus was intended to exhibit the process, it was
essential that an easy way should be provided for i^ctting away
the ingots and quickly repeating the operation. This casting
apparatus, constructed precisely as represented in Fig. 42, was
erected at my Bronze Manufactory in London, about two months
prior to my reading the " Cheltenham " paper, in August, 1856,
to which I shall refer later. The mold was 10 inches square, and
about 3 feet in length inside; it was made in two pieces planed
quite parallel, and then permanently bolted together. The
base was a massive square flange, resting on four dwarf columns,
which stood on the square upper flange of an hydraulic cylinder;
bolts passed through these dwarf columns, and through the square
flanges, thus uniting the ingot mold and hydraulic cylinder. To
the latter a ram or plunger was fitted, having a movable square
head, which accurately fitted the mold, and formed a movable
bottom to it. Both the ram and the external surface of the
mold were kept cool by a water jacket, provided with supply and
waste pipes. Matters being thus arranged, the converted metal
was allowed to fall in a vertical stream from the receiver on to
the head of the ram. The receiver was then removed and as
soon as the steel was solidified, water under pressure was turned
on to the h^^draulic cylinder, when a beautiful ingot, 10 inches
square and weighing about 7 cwt., steadily rose and stood on
end ready for removal, the head of the ram rising one or two
inches above the top of the mold. There are, no doubt, many
persons still living who witnessed this combined converting and
casting apparatus in successful operation.
Two I o-inches square ingots, made with this apparatus, were
sent to the Dowlais Iron Works in Wales, and, without hammer-
ing, were rolled into two flat-footed rails on the 6th of Septem-
ber, 1856; that is, twenty-four days after the reading of the
" Cheltenham " paper. They were rolled under the personal
superintendence of Mr. Edward Williams, past president of the
Iron and Steel Institute. Two pieces of these rails are still kept
at the Institute in a large glass case containing many other
examples of the early working of my process in London and in
Sheffield.
Before concluding this brief sketch of the earliest forms of
apparatus designed by me to facilitate or improve the process,
The Iron and Steel Magazine
The Genesis of the Bessemer Process 495
I must revert to the difficulties inseparable from a fixed con-
verter. In this form of apparatus much heat is dissipated by
the blowing which takes place during the running in of the metal
and by the continuation of the blast after the metal is converted
and during the whole time of its discharge, which is a period of
uncertain length. There is also the difficulty of stopping the
process if anything goes wrong with the blast engine, or if a
tuyere gives way. I searched diligently for a remedy for these
and other grave defects, which at that time appeared impossible
to remove, until the happy idea occurred to me of mounting
the converter on axis, so as to be able to keep the tuyeres above
the metal until the charge of molten iron was run in, thus per-
mitting the blowing of the whole charge to be commenced at one
and the same time, and admitting also of the cessation of blowing
during the discharge. This movement of the converter per-
mitted a stoppage of the process to take place at any time for
the removal of a damaged tuyere, if necessary, and afforded great
facilities for working.
The special form of the movable converter was also a matter
of great importance, and there were several requirements to
provide for. First, in order to make the heavy lining secure
when turned upside down, a more or less arched shape in all
directions was necessary. A long oval form seemed best adapted
to the purpose, as it allowed some eight or nine feet in height for
the metal to throw itself about in without leaving the converter.
Then the large mouth or outlet pointing to one side was desirable,
so that the sparks could be discharged away from the casting pit.
After much study I arrived at the form shown at A, Fig. 43,
Plate XV, which is an external elevation; B is a vertical section
showing the position in which the vessel is retained during the
running in of the metal; C shows it during the blow, and D the
position it assumes when the converted metal is poured into a
loamed-up casting ladle. This ladle is shown at E and F; it is
provided with a discharge valve at the bottom, so that it can be
moved from mold to mold by closing the valve during such move-
ment, and then permit a vertical stream to descend into the
mold perfectly free from any mixture of slags. The advantage
of this mode of filling the molds will be understood when it is
borne in mind that they are necessarily narrow unright vessels.
It is well known that a stream of molten metal poured from the
496
The Iron and Steel Magazine
PLATE XVI
2
CD
CD
>
O
!-i
a
ID
u
o
J Jic Ci diesis of the J^csscnicr Process 497
PLATE XVII
498 The Iron and Steel Magazine
lip of a ladle will describe a parabolic curve in its descent , tending
to strike the further side of the mold before reaching the bottom.
The surface of the cast-iron mold so struck is instantly melted
by the incandescent stream of steel, and the ingot and the mold
thus become united, causing great inconvenience. Nor is it
easy, in pouring the steel from the lip of the open ladle, to prevent
some of the fluid slag floating on its surface from flowing over
with the steel and spoiling the ingot. All of these difficulties
are avoided by the ladle fitted with a bottom valve discharging
a vertical stream down the center of the mold, the quantity and
flow being regulated with great facility by the hand-lever on the
side of the ladle. At G and H, Fig. 43, are shown the bottom
of the converter and the form of tuyeres.
Many other mechanical contrivances were necessary to per-
fect the process, such, for instance, as my patent blast engine,
with its noiseless self-acting valve ; the hydraulic crane carrying
the pouring ladle over every mold in the semi-circular casting
pit, and designed to rise and fall in accordance with the move-
ment of the converter when filling the ladle for casting; the
direct-acting ingot cranes, which clear the pit and refill it with
another set of molds rapidly, and with very little manual labor;
the elevated " valve-stand," from which safe position a single
workman can overlook the whole converting apparatus, and
control all their movements, govern the blast and work the
hydraulic cranes, etc.
The mode of transmitting semi-rotating motion to the con-
verter was another important problem which I had to solve. I
was of opinion that ordinary shafting and straps were inappli-
cable to this fiery monster. Five or ten tons of fluid metal had
to be lifted in one direction, this load diminishing until the fluid
running to the opposite end of the converter tended to reverse
the driving gear. If anything went wrong, or slipped, the con-
verter might swing itself round and discharge the incandescent
metal on to the floor or among the workpeople. These considera-
tions led me to adopt the hydraulic apparatus now universally
employed for governing the motions of the converter: for, with
this simple and reliable means, a lad at a safe distance can start
or stop it instantly, can alter its speed and motion and control
the pouring of a lo-ton charge with ease and certainty.
The first movable converter was erected at mv steel works
7 Jic (7ciicsis of the Hcsscnicr Process
500
The Iron and Steel Magazine
T— I
+->
I {
§
u
«
o
The Goicsis of tJic Bessemer Proeess 501
at Sheffield and was moved by hand gearing, because at that
early date I had not invented the hydraulic apparatus just de-
scribed. This early converting plant did good work at Sheffield,
and was constructed precisely as represented in Fig. 44, Plate
XVI, which shows also the first modification of the hydraulic
casting crane, and its ladle with valve, afterwards elaborated
by me and rendered suitable for casting heavy charges of steel.
The development of this earliest form of plant is shown in Figs.
45 and 46, Plates XVII and XVIII, and Fig. 47, annexed. The
early experiments at Baxter House were so far successful as to
justify myself and some of my friends in entering into partner-
ship and erecting in the town of Sheffield, a steel works which
still remains in active operation under the style of " Henry
Bessemer & Company, Limited." These works were estab-
lished both for commercial purposes and also to serve as a pioneer
works or school, where the process was for several years exhibited
to any iron or steel manufacturers who desired to take a license
to work under my patents. All of these were allowed, either
personally or by their managers, to see their own iron converted
prior to their taking a license.
50 2 The Iron and Steel Magazine
DRY AIR FOR BLAST FURNACES*
TN the report in "The Iron Age" of July 27, 1905, pages
-■■ 214-216, of the proceedings of the Congress of Mining and
Metallurgy, held in connection with the International Exhibi-
tion at Liege, Belgium, a synopsis was given of the papers read
and the accompanying discussions. Several papers among those
read by title have not been published. From advance proofs
of these, furnished through the courtesy of the secretary of the
Metallurgical Section, Constant Renson, technical manager of
the Angleur Iron and Steel Works, near Liege, the following
abstracts have been prepared:
Effect of Dry Blast on Furnace Working
A. LODIN
The increase of 25 per cent in output, with a saving of 20
per cent in coke consumption through drying the blast, an-
nounced by Mr. Gayley at the New York meeting of the Iron
and Steel Institute last year, attracted much attention from
metallurgists. These results were obtained at the Isabella
furnaces, near Pittsburg, with a plant put up in accordance'
with Mr. Gayley 's last patent. Notwithstanding several obscure
points which were not cleared up in Mr. Gayley's paper at
the last May meeting of the Iron and Steel Institute, his com-
munication possesses great interest, owing to the practical
results obtained. The author considers that the only reactions
susceptible of being influenced by httmidity in the blast were
reduction of the silicon and manganese, with fusion of the pig
and slag. The water vapor introduced into the space above
the tuyeres reacts immediately upon the coke, while forming
carbonic acid and absorbing 3,220 calories per kilogram, or 5,976
British thermal units per pound of coke burned. As the pro-
portion introduced by the blast increases, the mean temperature
of the space above the tuyeres diminishes, the reduction of the
silicon first becomes less active and the pig has a tendency to
pass from gray to white. Then, on the temperature becoming
still lower, the charges cease to melt above the tuyeres, so that
* Papers presented at the Li^ge Mining and Metallurgical Congress,
abstracted in " The Iron Age," November 16, 1905.
Dry Air for Blast Fiiriiaccs 503
a scaffold forms. All absorption of heat occurring in the region
of the tuyeres, like that due to decomposition of the humidity
in the blast, will delay the descent of the charges and diminish
the production in a ratio corresponding with the variation, not
of the total quantity of heat disengaged above the tuyeres, but
of the combustion temperature of the carbon when the normal
working is established.
If it be desired to maintain the output notwithstanding
this disturbing influence, a quantity of heat equal to that
abstracted must be returned to the zone of the tuyeres. For-
merly this could only be effected by burning an additional volume
of carbon, but heating the blast has afforded a far more effective
method of compensation, for it gives an exact counterpart of
the action exerted by the water vapor. The additional heat
introduced by the blast is utilized entirely by the fusion, and
its relative effect is so much the more considerable as the
combustion temperature of the carbon above the tuyeres was
originally lower, or as the working of the furnace must be kept
hotter.
It is impossible under the conditions described by Mr.
Gayley to regard the drying of the blast as the main cause of
the saving effected. It must contribute to reduce the coke con-
sumption, but if the part played by the absorption of heat
corresponding w4th the decomposition of the water in the blast
becomes reduced to almost nothing (as in one of the cases con-
sidered) another cause must be sought for difference in the fuel
consumption. The principal cause must be an influence acting
with constant intensity during the whole experimental period.
The diminution of one tenth the weight of blast passing
through the tuyeres during a given time is of a nature to exert
great influence on the working of the furnace. If the previous
working were too rapid it might be improved by a reduction
of the blast, which would prevent the ore from attaining the
space above the tuyeres in a state of incomplete reduction, so
as to become scarified there, while forming silicates that can
only be reduced by the action of solid carbon, with the forma-
tion of carbonic oxide and considerable absorption of heat. If
the descent of the charges should slacken, the reduction by the
gaseous current would be effected more completely, the coke
consumption would diminish and the composition of the furnace
504 The Iron and Steel Magazine
gases would be modified to advantage by increase in the ratio
of CO2 to CO.
It may be asked if the relative acceleration of working at
the Isabella furnaces was not obtained at the expense of fuel
economy, and if the improvement attributed by Mr. Gayley to
drying the blast is not chiefly due, during the winter period at
any rate, to reduction in the quantity of blast introduced in a
given length of time. For clearing up this question only one
of the factors capable of modifying the furnace working must
be varied at a time; and the experiments in this direction should
by preference be made on blast furnaces in normal working.
Everything leads to the belief that drying the blast would only
exert a subsidiary advantage if applied to furnaces blown with
a very hot blast and so arranged that the ore arrives in a reduced
state at the space above the tuyeres.
The Blast-Drying Problem
victor defays
It follows from the various communications made to the
congress on this subject that (i) drying the blast undoubtedly
reduces the coke consumption and increases the output; (2)
there is contraction in Mr. Gayley 's figures given at the New
York meeting of the Iron and Steel Institute, and not only so,
but (3) certain elements of appreciation are so far wanting that
the practical results cannot be verified by calculation; (4) M.
Divary's information (from experiments at Creusot, summarized
on page 404 of " The Iron Age " of August 17, 1905) as to the
influence of humid air on the working of his furnaces has great
interest because concording with Mr. Gayley 's results, while
also affording metallurgists a valuable method of investigation,
and (5) the proposed remedy will probably exert a different
effect according to local circumstances.
Considering that the question is of far greater importance
than is generally supposed, and thoroughly appreciating the
adage that for working well a blast furnace must have the feet
warm and the head cool, the author concludes: (i) A trial of
the Gayley system in any given blast furnace can only have
value for furnaces under the same conditions. (2) It is advis-
able to select twenty furnaces w^orking under the most different
Dry Air for Blast Furnaces 505
conditions possible and on which the Creusot observations
should be made for a given period, which would permit of de-
termining far more certainl}^ and with less expense than by a
simple trial the chances of advantage which metallurgists may
derive from drying the blast.
A furnace manager has no greater enemy than the steam
introduced at the base of his furnace. If a tuyere should leak
and not at once be attended to, the bottom of the furnace will
cool down and fire reach the mouth. What occurs on a large
scale in such an accident is always going on to a certain extent,
owing to the water vapor with which the blast is charged. In
fact, the water acts as a vehicle of heat units, which it conveys
from the bottom to the top of the furnace. As it appears well
established that the constant increase in the temperature of the
blast and the height of the furnace is chiefly for counteracting the
water vapor in the air, it may be asked whether on almost entirely
suppressing the cause there is not reason to return, to some
extent at least, to former practice? If the experiment be
successful it would naturally follow that very high tempera-
tures of blast must be abandoned and the profile of the furnace
must be remodeled, especially its height reduced.
Lower Blast Temperature
As regards reducing the blast temperature, the problem
of coke economy assumes a very different aspect now that waste
gases are utilized. The saving may take the form of one in
physical heat — that of the gas taken off from the mouth, a
radical loss, should be especially avoided, and all the more as
it complicates the purification of the gas — or there may be an
economy in the quantity of gas and its value in chemical heat
(improving in the ratio of CO2 to CO). If, thanks to drying the
blast, less gas be required to heat it, more will be left for other
purposes. And this diminution in heating the blast will have
another consequence, viz., that the stoves will be less costly,
which will go far to pay the expense of the drying plant. Per-
haps a return may even be made to types of stove that were
only abandoned because they did not raise the blast to a suffi-
cient temperature.
As to reducing the furnace height, its consequences, though
less important, would be appreciable, and at all events the old
5o6 The Iron and Steel Magazine
furnaces of moderate height now working would be in a better
position to compete with the modern giants.
Thus it will be seen that drying the blast may be attended
with important consequences and may reopen many questions
regarded as finally settled connected with the construction and
management of blast furnaces. The author's proposition to
make systematic observations on a large number of furnaces
working under the most various conditions as regards the
influence of the hygrometric degree in the blast would permit of
turning Mr. Gayley's invention to the greatest possible accoimt.
In a short time the amount of advantage that may be derived
from drying the blast, as regards coke saving, increased output
and regularity of working, would be definitely determined, while
accurate information would be afforded as to the type of furnace
which in each special case is best suited for turning these
advantages to the greatest possible account.
Utilization of Blast-Furnace Gas
victor defays
When it was found possible to utilize blast-furnace gas for
economically producing motive power, metallurgists regarded
this innovation as the dawn of a profound transformation. The
first trials seemed to show that the problem was solved almost
as soon as stated, and small scale experiments led to applica-
tions on a large scale. Difficulties encountered in freeing the
gas from fine dust brought a slight check, but soon inventors
found means of effectually eliminating dust from the gas, which
then became suitable for directly affording motive power.
In achieving this success, however, which greatly improved
the value of the gas as regards its utilization in engines, the
thermic qualities of the gas as a heating agent were also increased
so that the hot-air stoves and boilers fired with furnace gas
attained a far higher thermic efficiency than before. This
increased value of the gas for heating is due to two circumstances :
(i) Elimination of the greater portion of the sublimate which
adheres to the surfaces of the stoves and boilers, hindering the
transmission of heat by low conducting power. (2) The pres- .
ence of a quantity, always considerable, of water vapor in the
gas, which diminishes its heating power to a large extent.
Dry Air jor Blast Fiiniaces 507
Too Much Emphasis on Gas for Motors
Now that ironmasters possess a gas very much improved in
all respects, especially when freed of its humidity, there is reason
to ask whether the utilization of this gas, or at any rate some
of it, for firing metallurgical furnaces may not in many cases be
more economical than its exclusive employment for .generating
motive power. The author considers himself warranted in
replying affirmatively, although he does not pretend to solve
the problem which presents itself under many different aspects
according to the circimistances of each case. And the follow-
ing are his arguments in favor of this view :
1. Purification of the gas need not be carried to such an
extent for heating as for motive power.
2. There is a better coefficient of utilization in the former
than in the latter case, because in most iron and steel works
the demand vipon motive powder is variable, while there are
times when this demand greatly exceeds the mean; for instance,
when several large motors have to start simultaneously. The
utilization coefficient of gas for motive power in iron and steel
works may be put, according to circumstances, at from 40 to
70 per cent, if the blast furnaces themselves, with their blowing
engines and other accessories, be left out of the question. But
the question is very different with the firing of furnaces which
nearly all work continuously. Open-hearth furnaces, for in-
stance, consume day and night a constant quantity of gas, so
that the utilization coefficient easily reaches 80 or 90 per cent.
3. As regards the solid fuel burned, nearly all works com-
prising blast furnaces, steel works and rolling mills must con-
sume, in addition to the furnace gas, a certain amount of coal,
either for the production of motive power or for firing the fur-
naces. Blast-furnace gas for raising steam replaces coal of
inferior quality, but when it is used for firing furnaces it replaces
gas coal of higher intrinsic value, the gain in favor of the latter
utilization being put by the author at 12 J per cent.
Cost 0} Gas and Steam Power Plants
As regards first cost, it may be considered that the power
plant, whether gas or steam, costs about the same, all things
considered; but the balance is in favor of utilizing blast-furnace
5o8 The Iron and Steel Magazine
gas for firing the furnaces, combined with the production of
motive power by steam, if it be taken into consideration that
the plant for purifying the gas will cost less, and that, except a
few simple gas producers in reserve, it will be unnecessary to
erect these producers for the open-hearth reheaters and other
furnaces.
To sum up : In many cases there w^ill be appreciable advan-
tage in using the waste gases for firing furnaces instead of pro-
ducing motive power in engines, as is now the general tendency.
The slighter purifying of the gas, the better utilization coefficient,
the saving in the price of coal used to supplement the gas, and,
lastly, the lower first cost, are all in favor of this view. It may
also be advanced that steam engines at an electric generating
station are more supple and trustworthy, with lower cost for
up-keep, as compared with gas engines.
Producer Gas and Furnace Gas
Another paper enters into a comparison of producer gas
and blast-furnace gas for firing metallurgical furnaces, and con-
siders that if it be borne in mind that the quantity of carbonic
oxide is greater in the latter than in the former gas, and that
the former contains a great deal of hydrogen, the calorific value
of which at high temperatures is nil, or nearly so, and that there
is about the same quantity of nitrogen in both, they may be
regarded as practically equivalent for firing metallurgical fur-
naces. This will not hold good for gas engines, because, as their
temperature is far lower, hydrogen is a good fuel in their case.
It must be added in favor of blast-furnace gas that it may be so
far freed from water vapor as to become even dryer than pro-
ducer gas. In any case, if blast-furnace gas be slightly inferior
to producer gas for firing furnaces, the slight difference may be
very easily made good by the use of intensive burners, in which
a thin, flat jet of gas is enveloped by two similar jets of air,
and by a slight enrichment of the gas, which can be effected
very cheaply and simply.
As regards the point that has been raised of the radiation
of blast-furnace gas, the objection is unfounded that " furnaces
heat chiefly by reverberation, which requires a flame of great
radiation and consequently illumination, as is not the case with
blast-furnace gas, which contains scarcely any other fuel than
Dry Air jor Blast Furnaces 509
carbonic oxide, giving a n on -illuminating flame." Carbonic
oxide has great radiation, because Helmholtz has shown that
the non-illuminating flame of carbonic oxide has a radiat-
ing power superior to that of pure lighting gas as burned for
illumination.
The author concludes that at works where the blast furnaces
are near steel works and rolling mills with gas furnaces it will
be more economical to use the blast-furnace gas for firing these
latter and to raise steam for the motive power.
Gas or Electric Motor for Roll Driving
carl ilgner
The author classes roll trains in three categories, accord-
ing to the manner in which the power for driving them is utilized:
(i) Reversing mills; (2) trains always running in the same
direction, but at a speed varying from time to time in the ratio
of 3 to 2, while the slowest speed corresponds with the greatest
absorption of power; (3) those in which both power and speed
are constant. For the first of these classes the gas engine is
inapplicable, notwithstanding the attempts made at various
works to introduce a reversible coupling between the fly wheel
and the roll train.
Electric Drives Successjtd
Electric driving affords excellent means, not only for easily
and certainly regulating the speed, but also for transforming
the variable power required by the roll train into one that is
almost uniform, in which case there is no doubt of the success
of electric driving. Several electric roll trains are under con-
struction, and the preliminary trials hold out the hope of thor-
ough success. If a three-high roll train, the speed of which
varies periodically, be coupled with a gas engine, the latter will
furnish the greatest amount of power when the speed is lowest.
But the products rolled at this slight speed only constitute a
third of the whole output, while the remaining two thirds will
be rolled at the maximum speed of the gas engine, the efficiency
of which will be ver^^ slight. It follows that the gas engine will
give out a considerable portion of its power while consuming
too much gas per horse-power hour. By employing electricity
5IO The Iron and Steel Magazine
as an intermediary, the variations of speed are transferred to it
so that variation of the load on the motor is diminished. The
consequence is that, notwithstanding loss in the electric trans-
mission, and without regard to variation in the speed, gas
engines that are no more powerful than those for driving the
roll train directly may be erected at the generating station. It
is perfectly evident that reversing mills and cogging trains
absorb a widely varying amount of power. Electric driving
and centralizing the generation of energy present excellent
means for regulating motive power, on the one hand by increas-
ing the rotary masses in motion, to which are added those of
the fly wheels at the generating station, and on the other by
distributing the shocks and the irregularities over the whole
generating station. It is evident that the power thus required
by the cogging rolls from the gas engines at the central station
will be less than that which the rolls would absorb if each were
driven directly by its own gas motor.
Possibilities of Centralizing Power
As regards the third class — roll trains of constant speed,
which are generally used for plates and small bars — variations
of load are not considerable, and in their case electric driving
does not afford any great advantage. If, however, it be required
to drive roll trains of all three classes, there is no doubt that
centralization of the power is preferable to the use of a gas
engine for each separate roll train. And to the advantages
alreadv claimed for the electric driving of roll trains in classes
I and 2 must be added those resulting from centralization of the
power. The total power absorbed by all the roll trains at a
given moment is undoubtedly far less than the sum of the maxi-
mum power required by each train.
Another advantage is that the motor of the generating
station can receive all the care it requires, because one can be
kept in reserve. If the gas engine be coupled directly with the
roll train the stoppages required for overhauling will not be
compatible with proper working of the train. The electro-
motor stands overloading better than does the gas engine, while
more easily and at less expense it can be replaced by another
of greater power. The provision of a reserve for meeting hitches
at the blast furnaces or coke ovens is easier at a central station
Iron Resources oj the World 51 [
than for each nu)lor. In the hitter ease aU tliat can be done is
to hiv down i^\'is producers more difficult to manage, owing to
intermittent working. At the central station, on the contrary,
a steam turbine with a bank of gas-fired boilers constitutes a
tnistworthy reserve.
In short, a whole series of weighty considerations count in
favor of centralization. If it be considered that at the central
station much less power plant will need to be laid down, while
larger motors may be employed, and, if again, the connections
of each train and the long gas pipes be considered, the conclusion
is warranted that the first cost will not be an obstacle to adopt-
ing the principle of central electric stations. The difficulties
encountered in the progressive transformation of works must
not be disregarded, but it appears undeniable that centralizing
the motive power of rolling mills, with electric driving and the
use of blast-furnace or coke-oven gas, gives promise of great
economy as compared with present practice.
IRON RESOURCES OF THE WORLD *
By R. ANSPACH
pROF. A. E. TORNEBOHM, president of the " Sveriges
-*- Geologiska Undersokning," has made an exhaustive report
to the Swedish parliament, from which is taken the information
embodied in this paper.
In inquiring into the extent of the world's iron resources,
we must remember that, in the very nature of such an investi-
gation, it cannot be solved with any great degree of accuracy,
not even when only one country is considered, and still less for
a whole continent or the entire globe. For, on the one hand, the
location of ore deposits is known with tolerable accuracy only
in more or less civilized regions; while, on the other hand, it is
in many cases unknown how rich the various deposits may be
at greater depths. In addition to this, there is much uncertainty
as to what character of ore may properly be taken into account
in estimating ore supplies. For, besides the unquestionably
good ores, there are large quantities which are of less value, either
* Translated from the " Zeitschrift fiir angewandtc Chemie," " Engi-
neering and Mining Journal," October 7, 1905.
512 The Iron and Steel Magazine
owing to their composition or to their low iron content, or from
both causes. But there is a continual endeavor to find the
means for exploiting these ores of lower value, and the marvellous
development of the iron industry within the last twenty or thirty
years has been largely the consequence of progress made in
this direction.
To quote only a few familiar instances, the reader will recall
how the Thomas method for working up phosphorus-bearing
ores called to life the Norbotten fields, and the still far more ex-
tensive industry based on the minette (oolite) ores of Lorraine ;
he will further remember how the success recently achieved in
the magnetic concentration of ores has laid the foundation of a
gigantic undertaking for exploiting the low-grade, but other-
wise highly important, Dunderland ores in Norway. All these
ores were, before these improvements, regarded as almost un-
workable; and so the question naturally arises whether other
ores, which at the present day are considered practically worth-
less on other grounds, may not perhaps some day acquire im-
portance through further technical progress. Regarding ores,
belonging to this last class, data of any value are so meager, so
far as foreign deposits are concerned, that it will be necessary to
leave them out of account altogether in these considerations.
On the Swedish iron ore fields Professor Tornebohm has col-
lected the following material, for which absolute accuracy can-
not be claimed, as was pointed out above, and which is based in
part on rough estimates.
Norbotten. — According to the latest investigations the
ore supply in this district may be estimated as follows:
Kirunavara: Ore above the level of Luossa jarvis .... 265,000 ooo tons.
Ore beneath the same, down to a depth of 300 meters 510,000,000 ,,
Luossavara 18,000,000 ,,
Total 793,000,000 ,,
The ore is exceedingly rich; it contains 65 to 70 per cent
iron throughout, and the phosphorus runs high — i to 2 per cent
as a rule. Most of the ore (80 per cent) is exported to Germany..
In England there has as yet been little demand for the ore, as.
only a few iron works have taken up the basic method (Thomas
method), by which it has become possible to produce good iron
from phosphorus-bearing ores. But, according to recent in-
Iron Resources of the World 513
formation, this method is finding more and more application in
England also.
Gellivare. — The ore supply in Gellivare, abovethe railway
line, is estimated at 53 .800,000 tons and down to a level 100 meters
below this line at 49,700,000 tons; altogether, therefore, at 103,-
500,000 tons. But as ore is undoubtedly still found at greater
depths than that indicated (at any rate, in the more important
mines), we may, without danger of overestimate, add a quantity
of at least one half of that lying within 100 meters beneath the level
of the railway line — that is to say, in round numbers, 25,000,000
tons ; so that the total ore supply in Gellivare may be estimated at
128,500,000 tons. The iron content of the ore is 55 to 65 percent;
the phosphorus varies widely, but is considerable throughout.
The most important of the remaining Norbotten iron fields
are: Ekstromsberg, Mertainen, Svapavara, Tuolluvara and
Levaniemi. The quantity of ore in Ekstromsberg is estimated
at about 100,000,000 tons; that in Mertainen and Laukujarvi at
about 5,000,000 tons. The percentage of iron in these two fields
is 55 to 65; the phosphorus is rather high in Ekstromsberg, low
in Mertainen. There are no estimates available of the other
three fields, but they can be roughly gauged according to the
known area of the ore deposits. This area is :
Svapavara, 50,000 square meters 12.355 acres
_ Tuolluvara, 10,000 ,, 2.471 ,,
Levaniemi, 40,000 ,, 9.884 ,,
Total ....100,000 ,, 24.710 ,,
Supposing these fields to be worked to a depth of 200 meters,
and reckoning 3.5 tons of ore to the cubic meter, gives 70,000,000
tons as the total quantity of ore. The percentage of iron
in this field is 60 to 70; the phosphorus is comparatively low
in Tuolluvara, while in the other districts it varies a great deal,
though it is high throughout. The stock of ore in the chief iron
fields of Norbotten, is, therefore:
Kiruna — Luossavara 793,000,000 tons
Gellivare 128,500,000
Ekstromsberg 100,000,000
Mertainen Laukujarvi 5,000,000
Other mines 70 000,000
Total 10 ;6, 500, 000
514 The Iron and Steel Magazine
Central Sweden. — The ore supply of Grangesberg, down to 300
meters below the surface, is estimated at 60,000,000 tons. The
ore supply in the numerous other iron mines of central Sweden
can, at the present time, be estimated only roughly, according
to the area of the fields. This may be taken to be about
200,000 square meters (49.42 acres). As the more important
mines are already much attacked, we may assume that they can-
not on an average be worked further down than for another 100
meters. Assuming that each cubic meter yields 2.25 tons of ore, the
total quantity of ore would be 45,000,000 tons. On this basis the
total ore supply for central Sweden is, therefore, 105,000,000 tons,
and for the whole realm, in round numbers, 1,200,000,000 tons.
Two important ore deposits — Routivare in Norbotten and
Taberg in Smaland — have not been taken into account above,
as their ore is highly titaniferous, and, therefore, does not reach
the market. For Routivare the area of deposit is given as 300,-
000 square meters (74.13 acres). This, however, rather represents
collection of nodules of ore than a continuous deposit, and it is,
therefore, uncertain whether the depth corresponds at all to the
area of the district. It is at present impossible to estimate the
quantity of ore.
Taberg has an area of about 260,000 square meters (64.25
acres). The ore body is more collected here and probably has a
considerable depth, but the iron content is low^ (in the richest parts
30 to 40 per cent) ; this, together with the high percentage of
titanium (5 to 6 per cent), has hitherto prevented the exploita-
tion of the ore.
Foreign Iron Deposits. Norway. — In recent years several
important deposits of iron ore have been discovered in the north-
em part of Norway. The most important are the Dunderland,
the Naeverhaugen and the Sydvaranger districts. So far as the
area of the deposits is concerned, these much surpass the Nor-
botten fields, but the ores are, on the whole, poor (30 to 40 per
cent). Large installations for exploiting the Dunderland ores
are in the course of erection. The ores that can be reached by
surface working alone are estimated at 80,000,000 tons. It is
proposed to concentrate magnetically to 62 to 64 per cent, to
briquette the ore and to export annually to England 750,000
tons, there being a good demand for it there, owing to its low
phosphorus content. In Naeverhaugen and Sydvaranger no
I rati Resources of the World 515
estimates of any account have, as yet, been made of the ore
deposits. Statements regarding the quantity of ore differ a good
deal; the iron content varies between 30 and 58 per cent, but is
said to be 38 per cent on an average.
Outside of Scandinavia the countries which at present chiefly
produce iron ore in quantities affecting the world's market are
England, Lorraine, Spain, southern Russia and North America.
England. — The older English iron fields are now mostly
exhausted, and therefore abandoned. Most of the fields at
present under operation have been opened up within the last
ten years, as the Cleveland, West Cumberland, Lincolnshire,
Northampton, Derbyshire, Notts, Leicester and Oxfordshire
fields. The most important of these are the Cleveland fields, in
which, however, the iron runs low (about 30 per cent), and more-
over appears to be going down still further. In the year 1850,
when the Cleveland field was first opened, the ore supply ^as
estimated at from 4,000,000,000 to 5,000,000,000 tons. Since
then about 250,000,000 tons of the best ore have been extracted,
and what still remains of such ore will be exhausted in about
twenty years' time. The ore that will then be left is, in general,
of such poor quality that, with the present machinery and
methods, it would not be considered worth mining. The con-
dition in the other English iron ore districts is much the same as in
the Cleveland fields. The iron ore production of Great Britain
is decreasing. It reached its miaximum in 1882, with 18,000,000
tons; it is now about 12,500,000 tons, of which Cleveland con-
tributes 40.2 per cent; Lincolnshire and Northamptonshire,
26.7 per cent; Cumberland, 11.7 per cent; Scotland, 6.2 per cent;
and Staffordshire, 6.1 per cent. The annual consumption of
ore in England is, at the present time, 20,000,000 tons. The
deficit (some 6,500,000 tons) is covered by importation, chiefly
of Spanish ore.
Lorraine and Luxemburg. — The minette (oolite) ores which
occur here are of the greatest importance. They contain 35 to
40 per cent iron and 0.7 to 0.8 per cent phosphorus. The ore
supply is thus estimated:
German Lorraine 1,835,000,000 tons
French Lorraine 1,300,000,000 ,,
Luxemburg 300,000,000 ,,
Total 3,435,000,000 ,,
5i6 The Iron and Steel Magazine
The minette ores at present furnish 80 per cent of the total
production of Germany, and 66 per cent of that of France.
Spain. — The principal iron district of Spain is the Bilbao
field, on the north coast. The iron content of the ore there runs
from 50 to 55 per cent; the phosphorus is insignificant. The
original stock, once so extensive, is now greatly broken into, and
the production is decreasing. In the year 1899 it amounted to
6,5000,000 tons; in 1902 to 4,700,000 tons. This supply will
presumably be exhausted in ten or twenty years. The ore is
exported chiefly to England, which, of late 3^ears, has received
about 3,000,000 tons annualh^ In several places new fields
have recently been discovered in Spain, and mining has been
begun on somie of these, as for instance in Castile; in Asturia
(where are several deposits with an aggregate stock estimated
at about 200,000,000 tons) ; in ScAdlla Granda canal, 20,000,000
to 30,000,000 tons; in Paderoso, 10,000,000 tons; in Tornol,
50,000,000 tons; in Huelva (Cala), 18,000,000 tons; also several
deposits near the Mediterranean Sea, making a total stock of
between 50,000,000 and 60,000,000 tons.
Southern Russia. — There are here several important iron
fields, the two most important being Krivoi-Rog and Kertsch.
In the former the iron content is 50 to 65 per cent, the phos-
phorus as a rule less than o.i per cent. According to the most
recent statements, the stock of ore there is taken as 87,000,000
tons (older calculations gave much lower figures), which should
be exhausted in some thirty years' time. The production
amounted to about 2,500,000 tons in 1903, of which the greater
part was exported. The percentage of iron in the ore of the
Kertsch peninsula is 30 to 40, higher only in exceptional cases;
the phosphorus runs from i to 2 per cent. The stock is figured
at 846,000,000 tons, of which, however, only perhaps 13,000,000
tons have an iron content of 37 per cent or more.
North America. — The principal iron district in North Amer-
ica lies south and west of Lake Superior, in the United States.
On the Canadian side of the lake there are also a few iron fields,
but these are of comparatively small importance. There are
other iron fields in Alabama, Virginia and Tennessee. The
total production of iron ore in the United Sta.tes in 1902 was
36,000,000 tons, of which the Lake Superior field contributed
28,000,000, Alabama, 3,500,000, and Virginia and Tennessee,
Iron Resources of the World 517
1,800,000 tons. The Lake Superior ores are in part rich, with
55 to 60 per cent iron and 0.04 per cent phosphorus. The stock
of such ores has been quoted at 1,000,000,000 tons, but this,
according to more recent estimates, is too high. It is believed
that this supply will be exhausted before the middle of the
present century.
Until the year 1900 these mines produced exclusively ore
with a minimum of 60 per cent of iron. Since then the practice
of mixing the rich ore with poorer material has become more and
more widely established, so that the ore now produced does not
contain more than 52 to 54 percent of iron. The oldest of the
Lake Superior fields was opened in 1854, the youngest (the
Mesabi) in 1892. The latter is now the most productive (13,000,-
000 tons in 1903). Until 1903, inclusive, 253,000,000 tons had
been mined. The ores must be shipped to the blast furnaces
over very long distances (1,000 to 2,000 kilometers = 620 to 1,240
miles), mostly, however, by water. The ores in Alabama con-
tain 45 to 48 per cent iron, with a rather high percentage of phos-
phorus. There are coal measures near by. The known range
may be estimated at a low figure to hold from 50,000,000 to
60,000,000 tons, but probably the deposits extend considerably
beyond the field hitherto examined.
In addition to the ore deposits considered above, there are
still a large number of others, on which little or no mining has
hitherto been done. As the foremost among these should be
mentioned the iron ore range in the province of Shansi in north-
em China. In this district coal measures extend over at least
35,000 kilometers, with which iron ores are associated over a
large part of the region. For 2,500 years past the main portion
of China's iron consumption has been supplied from these ores,
but nevertheless the stock is only slightly attacked, and a very
extensive supply is still left over.
New ore fields have also been discovered in Ireland (county
of Antrim, calculated quantity of ore, 6,000,000 tons; iron content,
30 to 50 per cent), in the Cyclades, Algiers, the Soudan, Came-
roon, India, Tongking, Cuba, Peru, Mexico, New Mexico, Utah,
Oklahoma, Canada, New Caledonia, Western Australia, etc.
Regarding these deposits, however, no reliable information is
at present forthcoming.
The following table, showing the yield in 1901 of the iron ore
5i8 The Iron and Steel Magazine
districts of different countries, gives an idea of their relative
importance for the world's production:
Tons
United States 29,730,000
Germany (including Luxemburg) 16,840,000
England 12, 4;©, 000
Spain 8,050,000
Russia 5,gro,ooo
France 4,8; 0,000
vSweden 2,840,000
Austria 1,920,000
Hungary 1,660,000
Newfoundland 750,000
Greece 530,000
Algiers 520,000
Belgium 260,000
Italy (Elba) 240,000
Bosnia 130,000
Other countries 620,000
Total 87,500,000
The probable development of the iron ore situation in the
future can be summed up in the following paragraphs :
1. It can be foreseen with certainty that the iron ore fields
of North America, Germany and England will be exhausted
within one or two centuries from now; those bearing compara-
tively rich ore much earlier, even.
2. A decline, or the entire dying out, of the iron industry
in consequence would take place only in England, as in that
country the coal w^ill be used up about the same time. (It has
been calculated that the coal fields in Durham and Northumber-
land will run out in about 100 years, the other English coal fields
in from 250 to 300 years.)
3. In Germany and North America the deficit in the home
production of iron ore will be covered by importation, in accord-
ance with the well-known rule that the ores travel to the coal
fields, and not vice versa.
4. Apart from the countries in which industries are flourish-
ing at the present time, only northern China has the requisite
conditions for the development of an extensive iron industry, as
only there are coal and iron found associated together. Should
it, however, at some future time become possible, through tech-
nical advances, to recover iron from the ore with consumption
Shall We Substitute Iron for Steel ? 519
of little or no coal, a revolution in the state of affairs would take
place, the consequences of which can hardly be appreciated now.
5. The iron production in the coming century will in the
main depend: (a) On ores occurring in countries opened up to
industry at the present day, but having hitherto received no at-
tention, either on account of their low grade, or owing to their
being otherwise unsuitable ; (b) on the development of new min-
ing districts in lands as yet little touched by geological explora-
tion.
6. The location of the future centers of iron production will
be determined by the position of the coal fields, and by the con-
ditions of transport. These two factors, together with the
metallurgical progress in dressing the ores, will decide the coiirse
of the iron production in the future. The supply of ores to cover
the world's iron consumption will presumably never be ex-
hausted.
SHALL WE SUBSTITUTE IRON FOR STEEL?*
\ S many of our readers will recall, we published two years
"^^ ago articles from a large number of different contributors
describing experience with iron and steel pipe and with other
articles made of iron and steel, which showed apparently that
steel corrodes much more rapidly than iron. We present in this
issue considerable additional data from a discussion at the
Washington meeting of the American Institute of Mining
Engineers (the report of which has just become available for
publication). This discussion confirms the opinions previously
expressed and goes still further in showing the rapid rate of cor-
rosion of much of the steel made at the present day.
The Department of Agriculture at Washington is proceed-
ing with an investigation as the result of complaints by farmers
throughout the west, who declare that their wire fences, which
were formerly good for thirty years of life, now begin to fail by
the end of three years. Steel wire nails, steel plates, coated
with tin and used for roofing, steel boiler tubes and steel struc-
tural material are among the other products concerning which
complaint is heard because of the rapid rate of corrosion.
*" Engineering News," September 28, 1905.
5 20 The Iron and Steel Magazine
It is to be noted also that the importance of corrosion and
the limitation which it sets to the life of a structure are just
beginning to be fully appreciated by engineers, and in a lesser
degree by those who employ engineers. Forty years ago it
used to be considered that a wooden bridge Avas a temporary
structure and an iron or steel bridge a permanent one. But
there are wooden highway bridges still standing and in good
condition in this country which were old strvictures when steel
highway bridges first became common, and which have witnessed
already the final end of many of these same steel bridges. Of
course the great cause — and the one usually assigned — for
the replacement of an iron or vSteel bridge is that the traffic
has outgrown its capacity ; but it is also true in most cases that
corrosion, has greatly reduced the original strength. Who can
doubt that the engineers who Vjuilt the iron structtires that were
erected forty years ago, or at least those who paid for them,
would be disappointed to-day to realize how poorly their work
on the whole had withstood the ravages of time and how nearly
it had reached the limit of its life.
We at the present day have the advantage of experience
which they lacked. We can estimate closety — or ought to be
able to — a rate of annual depreciation, where they could only
guess, and it is an important question whether even more attention
ought not to be given to the matter of durability and resistance
to corrosion in the design of all sorts of engineering work.
Unquestionably this is already being done to a large extent.
The enormous use of concrete, and of reinforced concrete, and
the renaissance of the masonry arch, are evidence that dura-
bility counts for more than cheapness of first cost with many
users and their engineering advisers ; but unless all our informa-
tion is at fault we ought to go farther and return to the use of
iron for many purposes for which steel has become the custom-
ary material, or else find some method of making steel reason-
ably resistant to corrosive influences.
There are two things which greatly hamper the engineer
who wishes to employ iron instead of steel because of its greater
durability. He may write a specification calling for iron pipe
or nails or plate or sheets or wire, btit how can he tell whether
the material he receives is iron or steel without resorting to
some difficult and expensive chemical tests? And the retail
Shall ]Vc Sithstitittc Iron for Steel ? 521
dealer is no better off than the engineer, or as well off, perhaps.
In the seeond place, suppose the engineer does specify iron, can
he be sure that the iron he receives will be a satisfactory material ?
Iron fresh from the puddling furnace is used for selected brands
of boiler flues, stay-bolts, etc., but more or less of the wrought
iron of the present-day market is made from scrap. The con-
stitutents of the scrap pile from which that iron was created
are so varied and there is such large chance of steel forming part
of it and introducing the probability of non-homogeneity and
flaws that the engineer may be excused for hesitating before he
demands iron instead of steel.
Clearlv, then, the increasing use of wrought iron is closely
dependent on a reliable supply of wrought iron produced from
the ore, but the difficulties in the way of re-establishing the old
hand puddling system are many. A mere comparison of the
interior of an old-time puddling mill and a modern open-hearth
steel furnace plant will show that the one represents a maxi-
mum of hand labor, while in the other machinery, intelligently
controlled, has largely replaced human muscle. In order that
wrought iron may come into much more extensive use, some
process is needed by which wrought iron can be produced on a
large scale, with the minimum of hand labor, and with a prod-
uct equally reliable to that produced by the old method. A
process which appears to fulfill these conditions is that invented
bv Mr. Jas. P. Roe, of Pottstown, Pa., and described in a paper
published in our issue of May 7, 1903. The expert comments
on Mr. Roe's process made at the Washington meeting of the
Mining Engineers w^ere highly favorable and certainly indicate
that a way has been found whereby wrought iron can be pro-
duced at as low a cost as good steel.
Of course steel may be expected to hold its present place
in many fields and probably in most. Wherever hardness and
resistance to abrasion are important qualities, as in railway
rails, it is, of course, infinitely superior to wrought iron. Wher-
ever facility of welding or resistance to corrosion is an important
factor, the superiority of wrought iron must be admitted.
A question of much interest in this connection and one which
is still unsolved, relates to the cause of corrosion in steel. Cast
iron is known to be more resistant to corrosion than either
wrought iron or steel. In the early days of the Bessemer and
522 The Iron and Steel Magazine
open-hearth process it was naturally supposed that its product,
being intermediate in its carbon content between cast and
wrought iron, would be also intermediate as respects suscepti-
bility to corrosion. It was not until 3^ears of experience with
very soft steel had been obtained that the greater liability of
steel to corrosion was definitely determined. Even now it is
possible to find engineers who maintain that steel resists corrosion
as well as wrought iron. It is not long ago that there came to
our notice in the advertising literature of a wire manufacturer
the claim that the fence wire now made was as good as that ever
produced and that the reason the wire corroded so much more
rapidly than formerly was the greater amount of gas in the air
due to the increased consumption of coal in locomotive and
other boilers!
As to the real reason why soft steel corrodes so rapidly
engineers and chemists are, so far as we are aware, almost wholly
in the dark. It is, of course, quite within the possibilities that the
cause may yet be discovered and that it may then prove feasi-
ble to so alter the chemical constitution as to make the steel
longer lived. Certain alloys — nickel — for example, are known
to greatly increase the resistance of steel to corrosion ; but their
cost makes their extended use quite out of the question. Phos-
phorus is also believed to increase resistance to corrosion, but
its favorable influence in this direction is far more than counter-
balanced by its injury to the physical qualities.
These problems, however, must be left to the chemists.
Until they are solved the engineer who considers the durability
of a structure exposed to corrosive influence will prefer WTOught
iron to steel.
Great BriUu'ii's I rati uhhtstry 523
GREAT BRITAIN'S IRON INDUSTRY *
By T. GOOD
np HE future development of the British iron trades, upon
which the commercial prosperity of the country so largely
depends, is a subject not only of great interest, but of some
anxiety. New circumstances, new conditions of foreism
competition, have arisen; where once there were customers
there are now rivals. Great Britain must recognize that it has
no longer an absolute monopoly in any branch of industry ; that
in some directions it has already lost the leading position ; that
in others it is being more or less rapidly overtaken; and it is a
question whether the British are not laboring under some unfair
conditions of fettered enterprise and restricted effort against
which they cannot reasonably hope to struggle successfully in
the future as competition becomes keener.
It is not the object of the writer to take any part in the
fiscal controversy, nor to sound any note of pessimism; he
holds that there is nothing in this growing competition to be
afraid of, nothing to frighten either capital or labor, and that
there is no need to adopt any ill-conceived measures or doubtful
experiments in the interests, or alleged interests, of British in-
dustry, but he begs to urge the imperative need of cultivating
a more ready adaptability to new ideas, new times and new
circumstances.
Great Britain's proud position in the world of trade and
commerce was not attained without effort, nor will it be retained
without exertion. The time is here when it cannot well afford
to ignore a single commercial obstacle or industrial blight which
it is possible to remove. Wherever there is room for improve-
ment in method or policy, that improvement has got to be
made without undue delay, or prosperity has got to suffer.
And, with the possible exception of agriculture, in no direction
is there greater need for improvement than in the foundations of
the British iron industry.
Before offering any criticism of present methods, or making
any effort to forecast the future, let us take a glance backward.
Half a century ago Great Britain produced as much pig iron
* " Cassier's Magazine."
524 The Iron and Steel Magazine
as the rest of the world combined. In 1855 the output was
3,200,000 tons. From that time the actual, but not relative, out-
put steadily increased until 1872, when it produced 6,741,929
tons, valued at over ;^i8,5oo,ooo. Then the output declined
steadily until the production for 1879 fell to 5,995,337 tons,
valued at less than ;^i 5,000,000. Each of the next three years
showed a slight increase, — the production of pig iron in 1882
being 8,586,680 tons, of a net value of ;^24,042,704. During the
years 1883, 1884, 1885 and 1886 production and price fell, until
the year 1886 showed but 7,009,754 tons, worth ;^i5,888,775.
By 1889 it had improved up to an output of 8,322,824 tons
and a value of ;^2o,39o,9i8, while the following year, for a smaller
quantity, producers received over ;£24,ooo,ooo. Then both
quantity and value fell to 6,709,255 tons and ;^i 7, 276, 33 2 in
1892, and to ;^i5,898,445 for an output of 6,976,990 tons in 1893,
which was the year of the coal strike in the Midlands, — the
strike having commenced on July 28 of that year. During 1894,
1895, 1896, 1897 and 1898 there was a steady improvement.
In 1899 ^ record in quantity was established, and in 1900 a
record in price, the figures being: 1899, ;^32,66i,373 for
9,421,435 tons; 1900, ;^37,622,549 for 8,959,621 tons. Since
then the output has slightly decreased and the value has con-
siderably declined. In 1903 about 8,800,000 tons of approxi-
mately £22,000,000 value were prodticed, and in 1904 about
8,500,000 tons of about ;^2 1,000, 000 value. Table I gives the
British output of pig iron in the five-year periods from 1854 to
1900, and separately for the last six years.
Qiiantities in Quantities in
Millions of Millions of
Years Tons Years Tons
1855-9 3
1860-4 4
1865-9 5
1870-4 6
1875-9 6
1880-4 8
1885-9 7
1890-4 7
5 1895-9 8.7
2 1899 9-4
0 If; 00 8.9
3 I c o I 7.9
4 1902 8.6
1 1903 8.8
5 i(;o4 8.5
3
As recently as 1874 Great Britain produced 47 per cent of
the world's iron; in 1890 it produced but 27 per cent, and last
year not more than 17 per cent. As iron producers, the British
Great Bntaiii\^ Iroii hnlHstry 525
held first place until 1895, when the United States got ahead of
them; in 1Q03 they failed to retain even second position, —
being beaten by Germany. While the world's demand for iron
and iron goods has increased enormously, British production
has remained practically stationary. Whether we go back ten
years, or twenty years, or thirty years, we find that Great Brit-
ain's actual output of iron has not materially increased, while
relatively to the world's demand for, and comparatively to its
rivals' production of, iron it has lost ground to a very consider-
able extent.
While in iron production the British have been standing
still, and, per head of population, losing some ground, other
nations have been going ahead b}^ leaps and bounds. During
the last twenty years the French have increased their output
by about 30 per cent; the Belgians by 50 per cent; the Germans
by 300 per cent, and the Americans (United States) by about 400
per cent. Twenty-four years ago Great Britain produced 4^
cwts. of iron per head of population annually; to-day, when the
world is using double, if not treble, the quantity of iron it did
then, the British are only producing 4 cwt. per head. During
this decline from 4^ to 4 cwt. on per capitum basis, Germany
has increased from i to 3 -J-, and the United States from i J to 5^.
This proves that the British are losing their grip on the world's
iron market. This would not matter so much from a national
standpoint but for the fact that in another great branch of man-
ufacturing industry, viz., textiles, no progress relative to
increase of population has been made in Great Britain during the
last quarter of a century so far as the supplving of other markets
is concerned, while in agriculture there is shocking and shameful
lack of enterprise and progress.
And not only are the British being elbowed out of markets
abroad in which there is an ever-growing demand for iron and
iron goods, but they are being undersold in the home markets.
During twenty -five years, while British exports of iron and steel
and manufactures thereof have increased b}^ but a fraction per
head of population, their imports of these things have increased
by 300 per cent. Excluding iron ore and scrap (raw^ material),
the United Kingdom is now spending a round twenty million
pounds sterling per annum in foreign iron and iron goods, and
this while one third of their blast furnaces stand idle.
526
The Iron and Steel Magazine
p
w
b
H
(b
s
0
h-*
O
:?;
M
ffi
H
c
O
w
H
Q
2:
O
in
Oh
o
O
X
H
OT
K-t
PJ
l^
rt
^
D
z
H
a
m
<
'X
fa
\j
^
5 >
Q
<
W
CO
Q
<J
O
I— I
fa
o
o
u
o
ctS
.2 X
• t:; w
a ^
0)
^ fl C 03
w 03 03 ^
0)
to
cs
i-r; ro
t~~
O
«*
00
o
U-)
00
4 4
ro
6
0)
6
6
i
CO
-1-
O
"9
00 '^
01
o
ON
o^.
O
■T
oc
4 4
4
o
>-*
6
6
N
^ w
."1 >^
Ti-
^
Q g
o
O r^
'^
M
'-;
ro
W g
r^
lO Tt
4
IH
6
CN
M Pi,
i-O
vO OS
rO
.
•
U-)
t^
vo CO
4
N
cs
S Q
O ur,
« Q
r^
o o
O
M
fa Si
t^
lO CO
IN
4
Hi
fa O
O (X,
Pi §
^
ON •
•
VO
4
<N [
4
^ o
Tt-
m •
•
•
•
t^
ro
<^ '.
4
fa w
< >
r~»
On .
•
■
H
S
fO
H 1
4
Q
Z
<
M
!>.
•
•
o
W
CS \
ro
'p '^ "cu
>^ 1-1 .
U O w
QJ
.lU (Li
w cj
o C
L, W Wh
w 5 0) bjo
Ph 4:s '^ cu
c/2 ^ U H
Ph
o
<L»
o
w
P
■z
<
z
c
Pi
l-H
fa
o
o
Cu
o Lo q
o6 6 ^
-t -t O O)
o o
00
NC
r^
00
00
'1-
o
O
o
NO
NO
-t-
00
00
CO
•i-
O
o
^
NO
00
ro f^ 00
ro
M
f^ 00
00
O CO vo
--f-
o
O CJ
NO
ON
CO
o3
w
Ph
Ph
CD
O
X
ft ■ w
•c -d
•d
03
O
i-i
00 M
r^ ON
l-i lO
00 00
NO
6
o3 y
u o3
4
4
Ph
■d
Tl
!=l
c
03
03
1)
m
u
0)
O
d
-N (-H
fc- o3
bJO
w c/i u
o
o
^ ixi iu +->
w
0)
'P 1« OJ
O O 7)
03
03
u
03
o
O
r^H CJ H
o
'Ph
03
u
CD
tn
O
On
NO
o
O
o
ON
-d
03 •'-I
o
0)
0)
cu
a
(U
o
(U
OS
bO
o
Tl
:3
f )
Uh
^
(fi
*
Great Brita{)i\s Iron Industry 527
That America has some natural resources superior to Great
Britain will be readily admitted, although these are considerably
exaggerated from time to time. But the disadvantages under
which tlie British labor as compared with their American rivals
are much more artificial than natural, while in contrast to Ger-
many's progress, natural advantages cannot be urged in explana-
tion. The indisputable fact is that the great progress in the
production and manufacture of iron is primarily due, alike in
Europe and America, to low mining royalties, low transit rates,
and a ready adaptability to new inventions and ideas compared
with the conditions prevailing in Great Britain.
The problem of trade is the problem of cheap and rapid
production and transit, and whatever position Great Britain
has attained in the past when, in an industrial sense, other
nations were slumbering, it cannot reasonably expect to retain its
.supremacy in the face of keen competition if it continues to
tolerate the present excessive rents, royalties and transport
charges, and to pursue generally a too conservative policy in
industrial and commercial affairs.
The British stick too tenaciously to antiquated appliances
and ancient ideas. Take, for instance, blast-furnace practice.
The oatput per furnace in Germany is fully one third more, on
the average, than in the United Kingdom. With fully 50 more
furnaces in blast last year than had Germany, the British pro-
duction of iron was nearly 2,000,000 tons below^ that of their
rivals, w^hile in America, in 1903, 250 furnaces produced over
18,000,000 tons of pig iron as against 350 in Great Britain giving
an output of less than 9,000,000 tons.
The United vStates, having adjusted their recent financial
disturbances, are now producing pig iron at the rate of 22,000,000
tons a year, and there is evidence to support the belief that
American exports of iron and iron goods will shortly assume
proportions distinctly detrimental to the British. A good
number of American furnaces, with their modern equipments,
produce from 400 to 500 tons of pig iron daily, and an output
of 800 tons per day has been attained by special effort in at least
one case. An output of 300 tons per day is quite exceptional
in Great Britain.
The high -capacity furnaces of America, with their pig-iron
casting machines, whereby the work on the old-fashioned sand-
528 The Iron and Steel Magazine
bed is avoided, to mention but one point, have materially aided
the American industry. Then, again, the cost of labor both in
the mining and manufactiire of minerals is less in America than
in Great Britain. Electric coal-cutting machinery, and other
mechanical appliances in the mines, at the coke ovens and
smelting furnaces, are much more extensively used than in the
United Kingdom. The utility of these appliances is at once
apparent when we note that against the British output of 275
tons of coal per year per man employed in and about the mines,
the output in America is 526 tons. There are over 7,000 coal-
cutting machines at work in the United States, and they have
reduced cost by from 15 to 30 cents per ton.
However, the British manufacturer has quite a plethora of
candid critics continually tendering advice gratis, and advice
at so much per column, and while, as a practical man, the
writer of this article is well acquainted with the manifold short-
comings of the average British manufacturer and workman, he
contends that there is another audience — the landlords — to
which the critics of British industrialism should address some of
their strongest remonstrances.
A great deal has been heard within recent years of land
bills and fair rents courts for Ireland ; it is time a move in this
direction was made on behalf of the industries of Great Britain.
It is an indisputable fact, blink it as we may, that the unfettered
and unique land monopoly of the United Kingdom constitutes
the heaviest burden now suffered by British industry and is the
cardinal point in the problem of foreign competition. Hitherto,
the British have prospered in spite of the land laws simply be-
cause they had no formidable competitors in manufactures, but
now that other nations are developing their mineral resources,
manufacturing for themselves and for export, copying the best
British methods but discarding the worst ones, assisted by
enlightened land laws and transit facilities, the time is at hand
when the British must set about in earnest to mitigate the evils
of rents, royalties and railway rates excessively burdensome
compared with those enjoyed elsewhere.
Cheap transport charges and low rents and royalties have
been the primary factors in the rapid industrial development of
Germany and America, and the British cannot reasonably hope
to hold their own against these rivals unless they very materially
Great BriiiU)i\s Iron hidusiry 529
reduce their cost of carriage and their mining royalties. Some
ten years ago the Iron Trades Delegation to the Continent, upon
their return, gave it as their opinion that if British ironmasters
enioved the same mining royalties and railway rates as those of
their rivals, foreign competition in neutral markets could be
defied. And there was, and is, ample evidence to justify such
a contention.
From a geographical standpoint the transit facilities of
Great Britain should be superior to, and their cost cheaper
than, those of their rivals. But as a matter of fact,-— thanks to a
monopoly that has charged the railways vSuch scandalous prices
for land that they are saddled with capital charges twice and
thrice as heavy per mile as those of other countries, — the
British railway rates are notoriously oppressive compared with
those prevailing elsewhere, while the canal system is so anti-
quated and inefficient as to constitute a positive disgrace to an
industrial community.
What has been done in the matter of throwing heav}^ capital
charges upon the railways cannot very well be undone, but by
insisting upon having land at reasonable prices for the further
extension of railways, by more intelligent cooperation amongf
traders and between traders and railway companies, and by a
much-needed development of the waterways, the cost of carriage
could be materially reduced and the trade of the country ma-
terially increased. As it is, the railway rates for iron goods
from works to port, distance for distance, are on the average 115
per cent higher than those of Germany, 120 per cent higher than
those of France, and 300 per cent higher than those of Belgium
and America, while cases could be quoted where on the railways
of America and on the splendid canals of Germany the cost of
carriage for iron ore and other minerals is but one sixth of the
cost in the United Kingdom.
Another grievance which calls imperatively for redress if
Great Britain is to remain a great iron manufacturing nation
is in the matter of mining rents and royalties, — another phase
of the land monopoly. In the various charges termed fixed
rents, dead rents, lease-fees, wayleaves, waterleaves, airleaves
and royalty per ton of minerals raised, there is a list of extrava-
gant, exorbitant and extraordinary tariffs levied unjustly upon
British industrA?". These royalties — to group the various
53 o The Iron and Steel Magazine
charges under one heading, for they are all royalties upon, and
add to the cost of, mining — average not less than one shilling
per ton of coal.
This tax upon the production of coal, which is in turn a tax
upon the general industry of the country, and especially upon the
iron and steel trades in which so many tons of coal are used in
the manufacture of a single ton of finished goods, is a much more
serious matter than is generally recognized. On the most
moderate computation it is possible to make, the average
royaltv on a ton of British pig iron from iron stone, limestone and
coal, is not less than six shillings. And this pig iron with its
tax is but raw material for the manufacturers, further quantities
of coal having to be used in working it up. These mining roy-
alties are fair examples of how public well-being is sacrificed to
private privilege in Great Britain.
In Germany, mining royalties are fixed by the state at 2
per cent on the profits of the undertakings, and no prejudice of
the landowner is permitted to prohibit the working of minerals ;
in Belgium, at 2^ per cent on profits; in France, all coal and
ironstone being the property of the state, at 5 per cent on profits,
while rents are merety nominal, being about ^d. per acre, as
against British rents of from £2 to ;^5 per acre on top of the
lease-fees which are often excessive and sometimes prohibitive.
In Spain, also, mining rights are leased on nominal terms, and in
the United States royalties are practically unknown. While in
Great Britain mining royalties are a burdensome tax on pro-
duction, in the case of their Continental rivals they are only a
moderate charge on profits, and herein lies a solid grievance.
Now, suppose the British make a profit of two shillings per
ton in mining operations, or, what will serve the purpose of illus-
tration better, suppose no profit whatever is made, the royalty
charges will in any case reach about one shilling per ton. But
if the Germans make two shillings profit, their royalty will be one
halfpenny; if they make one shilling profit they will pay but
one farthing; while if they make no profit they pay no royalty.
But in the former case, whether profits are made or not, whether
the British can or cannot hold their ow^n in the matter of price
with competitors, the landlords' charges have to be met, or
operations must cease.
The whole system of granting mining leases in Great Britain
Great Britaiii's Iron Industry 531
is fundamentallv wrong, and is utterly opposed to industrial
welfare. A lease is conceded for a term of years upon conditions
mutually as^reed to bv tenant and owner, but when the lease
expires and the tenant company — their plant in full working
order and their business established — desire a renewal, it can be
obtained only on the landlord's terms, and if trade happens to be
good at the time, these terms are liable to he based on ''boom "
figures. Then, in the course of a few years, if trade slackens and
prices fall and the landlord is not of a generous disposition, the
inevitable collapse comes, — investors are ruined, workmen are
turned adrift and the plant becomes the property of the landlord.
This is no uncommon case. Quite recently the writer saw a
mine idle, the machinery in a state of decay and the cottages
deserted simply through excessive royalties which could not be
paid during a period of temporary depression. And this mine
is situated where all the natural conditions are favorable to the
industry; btit, as often happens, one man is permitted to set the
laws of nature and the needs of the community at defiance.
To further illustrate how these mining royalties handicap
British industry it may be mentioned that some iron manufac-
turers pay more in the shape of royalties than in wages, and that
the royalty charges on the bunker coal of a steamship amounts to
more than the crew's wages. The writer knows of one case — a
coal mine in Scotland — where the landlord's toll amounts to
35. 6d. per ton of coal raised. He knows of another mine in
Yorkshire paying no less than ;^4o,ooo a year in royalties on the
output of two shafts. He could quote the case of a firm paying
over ;£i 1,000 a year in royalties, but whose shareholders for nine
consecutive years were without a dividend; another firm paying
as much as ji^~ioo,ooo a year in royalties but nothing in profits for
several years; and yet another — this time a typical, not an
exceptional, case at all — paying ;/^8o,ooo yearly in royalties,
whereas a firm on the Continent with a similar output and making
a similar profit would not pay so much as ;,r3,ooo a year in roy-
alties.
In reference to raihvay rates and mining royalties taken
together, the writer will quote the case of a Yorkshire firm in
competition with a Belgian firm for an order from the London
County Council for rails. This firm uses Lincolnshire or North-
amptonshire ironstone upon which royalties amount to more than
53 2 The Iron and Steel Magazine
a shilling per ton, bnt to avoid exaggeration we will put it at one
shilling. It takes 4 tons of this ironstone to make one ton of
rails. This makes, in ironstone alone, 4 shillings per ton of rails
in royalty.
In the manufacture of this ton of rails there are used 3^
tons of coal. Taking the royalty upon coal to be only 6d. per
ton, to again avoid exaggeration, and reckoning the royalty
upon limestone to be only T^d., we have here, on coal, ironstone
and limestone, royalties amotmting to 6 shillings per ton of rails
as against the total royalty charges in Belgium of not more than
15. 2d. To take a ton of these rails from Leeds to London costs
115. 10^. Total royalty and railway rate, 175. loci. The Bel-
gian firm can deliver their rails in London for 55. per ton-carriage.
This leaves a balance in favor of the Belgian firm of ii5. 8<i. per
ton in cost of carriage and mining royalties.
While other nations are going ahead by leaps and bounds
in iron manufacture. Great Britain is standing still, — fettered
by a load of rents and royalties such as is not tolerated in any
other countr}^ Surely, then, it is high time to draw attention in
unmistakably plain language to this aspect of the problem of
foreign competition, — to this antiquated land monopoly which
is slowly, but none the less surely, undermining the very found-
ations of the British iron industrv.
O pi} I -Hearth Fuvuace Com pari so us 533
OPEN-HEARTH FURNACE COMPARISONS *
By A. D. WILLIAMS, Jr.
'T'HE proportions of open-hearth furnaces are usually arrived
^ at by an empirical ratio with their normal capacity in tons
of steel produced per heat. While it would be feasible to con-
struct rational formulas for proportioning the various parts of
the furnace, the labor involved in using such formulas would con-
sume a great deal of time and the results would be no more
accurate than those arrived at by the empirical ratios used.
Past experience alone will enable the designer to predict fuel
consumption and output within limits when the furnace is used
for the line of work for which it was designed, but radical changes
in the charge and the method of working will produce corre-
sponding changes in the output and fuel consumption.
Improvements came Slowly
For a long time furnaces were built without any considera-
tion of the local conditions. A furnace was merely an assemblage
of brick and buckstays. In many cases an ill-considered design
was copied by the aid of a bricklayer, and while such a furnace
turned out steel the fuel consumption was high, as were the
other expenses of operation. The modern furnace builders
are studying the problems to be met, stimulated to a degree by
the progress made in this line abroad, particularly in Germany,
where small plants are the rule and rapid production is sought
after.
One of the first improvements made was in raising the roof
of the furnace, thereby increasing the size of the combustion
chamber and utilizing the radiant heat of the flame and at the
same time reducing the tendency of the roof to collapse at
unpropitious moments. The low roof sloping downward toward
the center of the hearth was designed to throw the flame directly
upon the charge, but it left very little room for the flame at the
beginning of a heat, and the natural result followed, which was
that the roof cut rapidly and ultimately collapsed. Additional
material added to the outside of the roof did not help matters,
and, finally, the lesson was learned that a thin roof with a suffi-
*" The Iron Age," September 21, 1905.
534
The Iron and Steel Magazine
cient amount of room beneath it to pass the flame not only lasted
longer but improved the action of the furnace.
Furnace Proportions
The early furnaces were small, and when the larger size
furnaces were designed, the proportions used in smaller furnaces
were often followed without much consideration being paid to
all the elements which should have been considered. The area
of the hearth was settled on the basis of a certain number of
square feet per ton and its length was made twice the width.
The volume of gas burned in a furnace is increased very nearly
in direct proportion to its capacity, and an addition to the
length is of more value than increased width, in that it allows a
longer time for the chemical combinations of combustion to
occur. The result is that the fuel is used to better advantage
and less of it is required. The following table gives the hearth
dimensions for a few furnaces and the ratio between hearth area
and capacity and the length and width :
Normal
Capacity. Length X Width.
No. Location Tons Feet
1 Pencoyd 70 30.00 X q.oo
2 Frodingham 100 32.00X12.00
3 Jones & Laughlin . .200 40.00X16.00
4 Pa. Steel Company . 50 32.00X10.00
5 Donawitz 30 27.00 X 10.00
6 Duquesne 50 27.00 X 14.00
7 Sharon 50 29.00 X 14-50
8 111. Steel Company . 35 21.87X12.25
9 111. Steel Company . 50 32.00X14.00
10 Wellman-Seaver ... 50 33.67X13.00
11 Wellman-Seaver ... 25 25.00X10.50
12 Shoenberger 35 24.00X12.00
13 Rechitza 10 13.50 X 7.54
14 Barrow 50 28.00 X 10.25
15 Unknown 50 29.00 X 15 00
16 Laughlin 50 30.00X1500
17 Homestead 40 26.33X12.50
18 Pottsville 40 24.00 X 12.00
earth Area.
Square
Feet
Per
Ton
Length
divided
by
Width
270
3.86
3-33
400
4.00
2.67
640
3.20
2.50
320
6.40
3.20
270
9.00
2.70
378
7-56
1-93
420
8.-10
2.00
267
7.62
1.79
448
8.96
2.29
438
8.76
2-59
262
10.50
2.38
288
8.23
2.00
lOI
10. I 2
1.82
287
5-74
2-73
435
8.70
1-93
450
9.00
2.00
329
8.23
2.1 1
288
7.20
2.00
The first three furnaces are used for the Talbot process.
No. 17, the Homestead furnace, is used for the Monell process.
The widths given for Nos. 3 and 4 are the average widths, these
Opoi-Hcarth Furnace Comparisons 535
furnaces being wider at the middle of the hearth and tapering
toward botli ends.
A cubic foot of molten steel weighs 430 pounds, and a ton
occupies about 5.25 cubic feet. From this it can be seen that
the average depth of the bath in the above furnaces will be frori
6 to 20 inches. In the Barrow furnace, No. 14, the depth of the
bath averages about 13 inches and the maximum depth is 26
inches.
Omitting the first three furnaces, which were designed for
the Talbot process, in which a deep bath is necessary and only
a fraction of the contents of the furnace are drawn off at a time,
the above table can be summarized as follows :
( Maximum 10.50
Square feet of hearth pert on -I Minimum 5.74
( Average 8.29
i Maximum 3 -20
Ratio of hearth length to width < Minimum 1.79
( Average 2.23
Most of the American furnaces run under 1 5 heats per week
of approximately 135 hours. Occasionally as many as 23 heats
have been made, but such records are exceptional and not sus-
tained. In Europe furnaces making 23 heats per week do so in
regular practice and keep it up, but according to American stand-
ards these furnaces are operated under their capacitv — that is,
there is a great deal more furnace used to produce a ton of steel.
A Limit to the Width
In building open-hearth furnaces there is a limit to the
width of hearth practicable, owing to the limit to the strength
of the skew backs and the arch brick adjacent. It is also more
trouble to make and patch the bottom in a wide furnace than it is
in a narrow one. A long furnace with the usual arrangement
means a considerable addition to the length of the building
required to house the plant. The building is UvSually designed
to suit a certain furnace, and radical changes in an existing
plant are difficult to make owing to the surroundings, building
columns, etc.
Proportioxtxg the Regenerator Capacity
While the hearth area has a certain effect on the rapidity
with which the furnace works, the regenerator capacity must be
536
The Iron and Steel Magazine
so proportioned that the gas and air are at a proper heat on
reaching the ports so that they will combine without withdrawing
heat from the furnace. The height is really the most important
of the checker dimensions, as when the volume is the same the
higher checker will give better results than the lower. A great
many open-hearth plants are built on the banks of rivers, where
the general level of the plant is too close to the ground water line
to permit of the construction of vertical checkers, and the hori-
zontal s^^stem, which is not desirable, has been used.
The following table gives the dimensions of the portion of
the regenerator chamber occupied by the checker brick in a few
of the furnaces cited in the preceding table. The number refers
to the preceding table, and the capacity of the furnace has been
repeated to avoid referring back:
Capac-
ity.
No. Tons Length X Width X Height
6 50 Air, 22.00 X 10.00 X 9-00
Gas, 22.00 X 6.00 X goo
8 35 Air, 12.17X 8.33X 8.50
Gas, 12.17X 6.33X 8.50
Q 50 Air, 22.00X10.83X10.50
Gas, 22.00X 7.92X10.50
10 50 Air, 23.50X 8.00X10.00
Gas, 23.50X 5.50X10.00
11 25 Air, 18.00X 6.00X 8.50
Gas, 18.00 X 4-50 X 8.50
14 50 Air, 12.25X 9-5oX 9.25
Gas, 12.25X 9-5° X 9.25
15 50 Air, 18.00 X 12. ooX 14.50
Gas, 18.00X 8.00X14.50
j6 50 Air, 18.17X12.08X14-00
Gas, 18.17 X. 10.08 M 14.00
Area,
Square
Feet
Area di
vided
by
Tons
Volume
Cubic
Feet
Volume
di-
vided
by
Tons
Air di
vided
by
Gas
220
4.40
1,980
39.80
132
2.64
1,188
23-78
352
7.04
3,168
63-58
1.67
10 I
2.89
859
24.60
77
2.20
655
18.72
178
505
1,514
43-32
^-31
238
4.76
2,499
49-98
174
3-48
1,827
36-54
412
8.24
4,326
86.52
1.36
188
3-76
1,880
37.60
129
2.58
1,290
25-80
317
6.34
3,170
63-40
1 .46
108
4-32
918
36.72
81
324
688
27.52
189
7-56
1,606
64.24
1-325
116
2.32
1,075
20.15
116
2.32
1,075
20. I 5
232
4.64
2,150
40.30
1 .00
216
4-32
3,132
62.64
144
2.88
2,068
41.36
3fio
7.20
5,200
104.00
1-52
2 19
4.38
3,066
61-32
183
3.66
2,562
5 '-24
402
8.04
5,628
112.56
J-I95
Opai-Hcarth Fniiiace Comparisons. 537
For each furnace the third hne gives the ratio for the com-
bined sum of the gas and air regenerators. No. 6 is fired with
natural gas, but is arranged to use producer gas if necessary.
Xos. 8. 9, 10, II and 14 are fired with producer gas. Furnace
Xo. 16 somewhat closely approaches foreign practice both in its
regenerators and hearth.
In some foreign furnaces the total volume of the regenerators
is 4 cubic meters, or 140 cubic feet, per ton, half of which is in
the gas and half in the air checkers. A more usual method is to
make the air regenerators 10 per cent greater in volume than the
gas regenerators, giving 74 cubic feet to the air and 67 cubic feet
to the gas checkers per ton, and in a few cases even larger regen-
erators are in use. Another ratio used abroad is to allow from
no to 155 pounds of checker brick per pound of coal burned in
the producers per ton of steel. One brick weighs 7.33 pounds,
and when set 3^ inches apart, which is the usual gauge for
checker brick, the above ratio would mean a checker volume of
105 to 150 cubic feet per ton, with a fuel consumption of 500
pounds of coal to be divided between the gas and the air.
In modern furnaces the regenerators are placed beneath
the charging platform and a good sized cinder pocket is pro-
vided, which catches the cinder carried over and prevents the
checkers from becoming bunged up and destroyed. The early
furnaces were supported on the regenerator casings and a great
deal of trouble w^as due to such construction, as this brick work
was continually expanding and contracting, racking the pan and
causing cracks in the bottom. Fortunately, in most cases these
cracks amounted to little, but whenever it was necessary to rip
out an old bottom it was found full of threads of steel, which in
many cases had reached the pan and chilled there without
causing a run out. Modern furnaces are supported independ-
ently of the flues leading to them, steel columns and beams being
used, upon which the pan rests, or in some cases a solid mass of
brick and concrete is used without a pan.
ABSTRACTS
#
(Front recent articles of interest to the Iron and Steel Metallurgist)
'HE Thermal Transformations of Carbon Steels. J. O. Arnold
and Andrew McWilliam. The Iron and Steel Institute,
September, 1905, meeting.
Ti,ooow., illustrated. — The
authors determined the ther-
mal critical points of three
samples of steel, containing
respectively 0.89, 0.21 and
1.78 per cent carbon and
small amounts of impurities,
and studied the corresponding
structural changes. The ac-
companying diagram repre-
sents their interpretation of
the nature and transforma-
tion of pearlite. The authors
conclude as follows:
'' I. The cooling transfor-
mations of an unsaturated
J. O. Arnold ^^.^^^ ^^^ carbon steel, previ-
ously heated to about 950° C, when studied under suitable
thermal conditions on a steel in which the points Arg, Arj
and Ar^ can be clearly differentiated, lead to the following
conclusions: Above Arg {i. e., 810° C), the ferrite and hardenite
are in mutual solution as a homogeneous mass. The Arg
'■' Note. The publishers will endeavor to supply upon request the f.:il
text of the articles here abstracted, together wiJi all illustrations, plans,
etc. The charge for this is indicated by the letter following the number
of each abstract. — Thus "A" denotes 20 cents, "B" 40 cents, "C" 60
cents, "D" 80 cents, "E" $1.00, "F" $1.20, "G" $1.60, and "H" $2.00.
Where there is no letter the price will be given upon request. In all cr.ses
the article furnished will be in the original language unless a translation
is specifically desired, in which case an extra charge will be made depend-
i:ig upon the length and character of the text.
When ordering, both the number and name of the abstract should be
mentioned.
538
Abstracts
539
Shotctmj the Properties of Pearlite and its Decomposition Product.
Fe.fi represented Black,
Mechanical Properties of
Mass.
Microstructure.
Maximum tensile stress
about 70 tons per square
inch. Elongation on 2
inches=about 10 percent.
Maximum tensile stress
about 55 tons per square
inch.. Elongation on 2
inches = about 15 percent.
Maxin\um tensile stress
about 35 tons per square
inch. Elongation on 2
inches = about 5 per cent.
Segregation Stages.
1st Phase.
Sorbitic" [pearlite
emulsified Fe3C.
dark on etching.
with
Verv
2nd Phase.
Normal pearlite with semi-
segregated FejjC. Dark
on etching.
3rd Phase.
Laminated pearlite with
completely segregated
Fe3C. Exhibiting a play
of gorgeous colours when
lightly etched.
Maximum tensile stress
about 30 tons per square
inch.
4th Phase.
I>aniinated pearlite passing
into massive FegC and
ferrite.
Note. — It is important to remember that in a single section of steel two or even all
three phases of pearlite may be observed in juxtaposition gradually merging into each
other.
A 2
540 The Iron and Steel Magazine
change is accompanied by a segregation of the two constituents,
which, if the coohng be slow, is probably completed in the
Beta range of temperature. On reference to Micrograph No. 4
it will be seen that after a fairly rapid cooling from 950° C.
the 0.21 per cent carbon steel, when quenched at 730° C. (or
near the middle of the point Ars), micrographically registered
a segregation of ferrite so far advanced as strongly to suggest
that such segregation must have begun at Arg and not as Ar2.
In other words, hardenite is insoluble in ferrite in both the Beta
and Alpha ranges of temperature. It, however, still retains
its identity as hardenite whilst falling through, say, 30° or 40° C.
of temperature in the Alpha range — namety, from the end of
Ar2 at about 720° C. to the beginning of Ar^ at about 680° C,
at which latter temperature it begins to decompose into pearlite.
''2. The heating transformations of the above steel are
substantially as follows: At Ac^ (about 710° C.) in the Alpha
range the pearlite begins to change into hardenite ; hence the Beta
carbide is soluble in the Alpha range. The change to hardenite
is somewhat advanced when Ac^ merges into Aca at about
720° C, owing to these points always overlapping in the heating
curve. The hardenite areas probably remain unchanged on
the sites previously occupied by the pearlite till the Gamma
range ACg is reached (at about 810° C), when the hardenite
and ferrite dissolve into each other, forming a homogeneous
molecular mixture.
^^3. In a saturated steel there is, on heating, a single absorp-
tion of heat at the change point Aci,2,3, the amplitude of which
ranges from about 710° C. to 730° C. This change marks the
transformation of the whole mass from pearlite into hardenite.
^^ 4. On cooling, there is a very considerable evolution of
heat at the single point Ari,2,3, the amplitude of which ranges
from about 690° to 660° C. This recalescence marks the trans-
formation of hardenite into pearlite. The particular phase of
pearlite obtained depends upon the rate of cooling from 660° C.
to atmospheric temperature. The emulsified phase is produced
by very rapid cooling, normal pearlite by ordinary cooling and
laminated pearlite by very slow cooling. Pearlite, in which
the carbide is emulsified or ' sorbitic,' may also be produced by
tempering hardenite.
*' 5. The micrographic and thermal transformations of a
Abstracts 541
supersaturated steel are as follows: At Aci,2,3, the sectional
ground mass pearlite chanj^es to hardenite, the cementite
slowly segregates into larger masses until a temperature of about
()oo° C. is reached, then the cementite and hardenite dissolve
one into the other and a homogeneovis mass of hardenite and
cementite is obtained.
'^ 6. On cooling, at about 900° C. the cementite falls out
with a faint evolution of heat and is completely segregated long
before the point Ar^.o.g is reached; hence the micrographic
transformations of cementite and hardenite are quite uncon-
nected with the three thermal critical points or any of them
and are due entirely to the influence of temperature." No. 434.
The Cleaning and Agglomerating of Ore Dust. '^ The Iron
Trade Review," September 14, 1905. 2,600 w., illustrated.
— Description of the plant of the Ruthenberg Reduction Com-
pany, at Niagara Falls, N. Y., for the separation of impurities
from fine ores by magnetic action and the concentration of ore
dust for the blast furnace by an electric process.
To bring the ore dust into condition for practical use in
the blast furnace it must be converted into a shape which will
permit the gases of the furnace to percolate through the charge
and so must be of some size and weight. The product of the
Ruthenburg process is in the form of bean-like particles, and
constitutes an admirable form for use in the ironmaker. The
contrast in the phvsical structure of the concentrates before
and after agglomeration is shown by the accompanying repro-
duction from a photograph of an equal w^eight of concentrates
through a 20-mesh sieve and the Ruthenburg product.
To convert the concentrated dust, a tower which is similar
in appearance to a blast furnace cut off above the tuyeres, has
been erected and also constructed to provide for the reduction
of the ore at the same time, which is the third step in Mr. Ruthen-
burg's scheme. The tower, a sectional view of which, together
with the gas producer for reducing the ore, is shown, is about
thirty feet high and about thirteen feet in diameter at the base,
and built of brick. In the interior is an inner tower, between
the outside wall and which is an annular space, the purpose of
which will be shown later.
On top, over the mouth of the tower, is placed an electric
542 The Iron and Steel Magazine
furnace of Mr. Ruthenburg's design. The only inlet for charging
the dust into the tower is between the arms of the furnace,
which are electricall}^ insulated from each other and form the
terminals of a heavy melting circuit.
Current is supplied to the coils, and the two arms between
which the ore dust must pass into the tower are strongly mag-
netized. The dust is raised on an elevator which stands at one
side of the tower and fed directly on the two magnetic arms of
the furnace, assuming a condition as shown.
The resistance to the passage of the melting current, set
up by the fineh^ divided particles of ore, which form a mag-
netic bridge and an electrical resistance, heats them to the
point of fusion and melted globules and masses of the ore drop
to the bottom of the tower in a small and fairly uniform size,
which regularity is contributed to by the falling of the molten
particles through the distance from the top of the tower to the
level of the charge. This melting is continued at the rate of
about 2,000 pounds per hour, using 165 kilowatts in that time,
which means, for the rate at which electric power is supplied at
Niagara, a cost of about fifty cents per ton. The tower is filled
up and then the process of reduction begins, the feeding of dust
and the drawing off of reduced metal sponge being continuous.
Tn an adjoining room, a hermetically sealed steel gas producer
has been erected for making producer gas. The object in her-
metically sealing is to enable the gas pressure, necessary to force
gas through the material in the tower, to be obtained by sending
air into the producer under pressure. To obtain a pressure of
five pounds, which is the blast used, the company has installed
a Connersville air blower driven by a three-phase alternating
current motor. The charge of coke is admitted through a de-
vice at the top consisting of a rotating barrel with two slots
less than 180 degrees apart. The charge is placed in the barrel,
which is then rotated, the charge falling into the producer as
the second valve opens downward, the charging valve then
being closed so that no gas escapes. The ashes are taken care
of by fluxing with blast-furnace slag. The hot gas under
pressure passes through the main containing a sand valve
into the bottom of the agglomerating tower where it forces
its way up through the ore. In the absence of the air the
ore is reduced by the gas, which, in turn, is converted in part
Abstracts
543
into carbon dioxide and on reaching the top burns in contact
with air and is diverted downward into the annular space sur-
rounding the inner tower, thus aiding in maintaining the heat
of the ore. The gas is exhausted near the bottom of the tower
into the open air through a chimney. From the time that any
given part of the ore is charged into the tower until it reaches
the bottom and is ready to be drawn out as wrought iron through
ports provided, about twenty -four hours elapse. The sulphur
2<.'t^iT7^l'
Jt^''=^/£u;^
in the ore is eliminated in the furnace during the fusion of the
dust, when the following reaction takes place: 2Fe304 + 4FeS
= 5Fe2 + 4SO2. Concentrates with as much as 0.8 per cent of
sulphur give an agglomerated product with 0.03 per cent sulphur.
The SO2 gas can be readily seen coming from the furnace at the
top of the tower, and as there is no sulphur in the producer gas
the iron taken from the bottom of the tower is practically free
from that impurity. No. 435. A.
I90S'
544
The Iron and Steel Magazine
The York Process for Rolling Steel Ties from Old Rails.
The Railroad Gazette," November 24, 1905. 2,000 w., ilhis-
FlG. I.
trated. — The York process of rolling is a radical departure from
any previous rolling-mill practice and enables almost anv desired
Fig.
Fig. 3.
section to be made from'
either the head or bot-
tom flange, no matter
how badly worn or un-
symmetrical the scrap
stock may be. Fig. i
shows a few of the many
structural shapes which
can be rolled from the
rail shown on the left.
Figs. 2 and 3 show the
forms of ties which may
be rolled, one with a fiat
bottom and the other
Abstracts 545
with a concave bottom, givinij: the tie itself elasticity in the
ballast under heavy loads. No. 436. A.
The Kjellin Electric Steel Furnace. '' The Iron Age,"
October 19, 1905. 5,000 w., illustrated. — A description of the
construction and operation of the Kjellin Electric Steel Furnace,
with special reference to the steel works at Gy singe, Sweden.
No. 437. A.
Fortschritte im Bau von Gasofen fiir Eisenhuttenwerke
(Improvements in the Construction of Gas Furnaces Used in Steel
Works). A. Desgraz. 3,000 w. '' Stahl und Eisen," July i,
1905. No. 438. D.
Die Herdofenstahlerzeugung aiis Flussigem Roheisen (The
Open-Hearth Process for the Production of Steel from Cast Iron).
Oskar Simmersbach. " Stahl und Eisen," June 15 and Jtily i,
1905. 3,800 w. — The author describes the growth of the open-
hearth process, comparing it to that of the Bessemer process.
No. 439. D-
Experimentelle Studien iiber die Vorgange im Hochofen
(Experimental Studies of the Reactions in the Blast Furnace).
F. Zimmermaim. " Stahl und Eisen," July i, 1905. 2,000 w.
— The author discusses the reactions taking place in the blast
furnace as indicated chiefly from the composition of the gases
taken from various zones of the furnace. No. 440. D.
. The Presence of Greenish -Colored Markings in the Fractured
Surfaces of Test Pieces. H. C. Howarth. Iron and Steel In-
stitute, September, 1905. 7,000 w., illustrated. No. 441.
Wear of Steel Rails on Bridges. Thomas Andrews, Iron
and Steel Institute, September, 1905. 10,000 w., illustrated.
No. 442.
Experimental Desulphurization. Reginald Meeks. '' The
Iron Age," November 9, 1905. 2,000 w. — The author describes
some attempts to reduce the percentage of sulphur in cast iron
(i) by the addition of much limestone to the charge; (2) by the
use of manganese ore and (3) by the addition of ferro-manganese
in the cupola or in the ladle. The f^rst two methods were found
ineffective, while the use of terro-manganese res'ilted in a marked
decrease of the sulphur content No. 443. A.
METALLURGICAL NOTES AND COMMENTS
Standard Method for the Determination of Silicon in Cast
Iron. — At the annual convention of the American Foundrymen's
Association held in New York City, June 6 to 8, the following
report of the metallurgical section read by H. F. Dilles, secretary,
was submitted and duly adopted :
'^ During the past year your committee has formulated a
method for determinating the silicon in cast iron, and is now at
work on the question of the total carbon. The following is the
method which your committee recommends to be the standard
of the association, for the determination of silicon in pig iron
and cast iron :
^^ ' A¥eigh one gram of sample, add 30 cc. nitric acid (1.13
sp. gr.); then 5 cc. sulphuric acid (cone). Evaporate on hot
plate until all fumes are driven off. Take up in water and boil
until all ferrous sulphate is dissolved. Filter on an ashless filter,
with or without suction pump, using a cone. Wash once with
hot water, once with hydrochloric acid, and three or four times
with hot water. Ignite, weigh and evaporate with a few drops
of sulphuric acid and 4 or 5 cc. of hydrofluoric acid. Ignite
slowly and weigh. Multiply the difference in weight by 0.4702.'
" In recommending the above method, it was recognized
that it is almost an impossibility to get chemists to use a stan-
dard method in their daily work. Hence the above method, as
recommended, is intended primarily as a check method in case
of dispute between different laboratories, or as between buyer
and seller.
" Hence a method accurate in every point was sought,
shortness being sacrificed to some extent to insure accuracy or
the chance of error by a careless operator. Little in the above is
left to the judgment of the chemist.
" It will be further recognized that in the purchase and sale
of pig iron or castings under specification, that standard methods
are essential in order to allow the parties of both parts to make
546
Metalluriiical Notes and Comments
547
their determinations with the assurance that, on the score of
method, they are on the same footing."
A New Method of Preventing Pipes in Large Ingots.* — In
the manufacture of large steel ingots for forgings or other pur-
poses it is often necessary to allow for a discard as high as 25 or
30 per cent of the total weight on account of the " pipe " formed
as the metal contracts in cooling. The use of a sink head lined
Fig. I. End Elevation
with fire clay or other refractory material, for the purpose of
keeping the top of the ingot longer molten, is successful to a
certain extent and results in a shorter pipe, but does not alto-
gether eliminate the piping. Covering the molten metal with
charcoal or similar material has the great disadvantage that a
considerable carburization of the upper third of the ingot often
* From F. O. Beikirch, in " Stahl und Eisen." " Iron Age," October
5. 1905-
548
The Iron and Steel Magazine
results, while if slag or sand is used a portion is often drawn into
the interior. The use of hydraulic pressure to compress the steel
while passing from the molten to the solid state has the desired
effect, but the cost of installing and operating machinery for this
purpose is in most cases prohibitive.
The process here described, which in most countries is pro-
tected by patents, has been in use for a year at the Gutehoff-
FlG.
Side Elevation
nungs Works in Oberhausen, Germany, giving good results on
ingots up to 60 tons weight. It is based on the theory that ex-
ternal heat is necessarv to keep the steel in the sink head molten
until the ingot is solidified and all danger of piping has passed.
This heat is obtained by blowing cold air through incandescent
coke, so regulating the pressure that in the fuel chamber only
carbon monoxide is formed, combustion to carbonic acid being
completed in the space above the sink head.
Metallurgical Notes and Comments
549
The original article reproduces photographs of four ingots
cast bv this method, which show ahiiost complete freedom from
pipes. The weight and amount of discard necessary with each
of these ingots were as follows: No. i, weight it. 6 gross tons,
discard 7.3 per cent; No. 2, weight i 7 tons, discard 4.98 per cent;
No. 3, weight 17.2 tons, discard 5.52 per cent; No. 4, weight 16.4
tons, discard 3.6 per cent.
The accompanying illustrations show the arrangement for
ingots from 10 to 60 tons. Fig. i is an end elevation, Fig. 2 a
side elevation and Fig. 3 the plan. The method of operating is
as follows: The chamber K is filled to the top with pieces of hard
coke about the size of a man's fist. About an hour before the
ingot is poured the fuel is brought to redness by means of a
Fi
G. 3.
Plan
gentle blast, the fiame which escapes at A being used to warm
the mold and more particularly the refractory lining of the head.
Shortly before the steel is poured the apparatus is drawn back
out of the way and at the same time the full pressure of blast is
put on, so that by the time the mold is full, in fifteen to twenty-
five minutes, the fuel is at a bright red heat, ready to be replaced
over the mold. As may be seen from the illustrations, the ap-
paratus is placed on a carriage, which can cjuickly and easily be
moved forward or back. As cold blast is used the blower can be
placed close to the casting pit, so that the whole arrangement
is very simple and compact.
55© The Iron and Steel Magazine
Iron Ore Analysis at Lake Superior Mines.* — The idea of
compiling the methods used in the laboratories of the iron mining
companies of the Lake Superior district in the analysis of iron
ores was suggested by the work of Francis G. Phillips on " Meth-
ods for the Analysis of Ores, Pig Iron and Steel," published by
the Engineers' Society of Western Pennsylvania, 1896, and later
in book form by the Chemical Publishing Company. The plan
therein carried out has been followed in the present inquiry.
A letter was addressed to each of the chemists whose name and
address the writer was able to obtain, requesting a description of
the methods used in the analysis of iron ores in the determination
of iron, phosphorus and such other substances as might be deter-
mined. The responses were quite general, but few^ of those ad-
dressed failing to send in a description of the methods employed.
The intention is not to present a scientific treatise on iron
ore analysis, but to set forth in some detail the methods of pro-
cedure carried out in the daily work in the analysis of iron ores,
which furnishes the basis for the grading of the ores and the
commercial transactions of the mining companies. Because of
the limited time allowable for analyses, and the accuracy and
reliability that results must present, we believe that in the Lake
Superior district the methods employed are as rapid and at the
same time as reliable as may be found in use in any section in the
commercial analysis of iron ores. That the present compilation
is in many respects crude and imperfect we are well aware. The
attempt was made to reach all the chemists of the region and the
desire was to give all an opportunity to contribute to the work.
No doubt some may have been missed.
The reports coming from the chemists of the different mining
sections are distributed as follows: From Minnesota five, repre-
senting the Vermilion and Mesaba ranges; from the Gogebic
range six, from the Menominee range five, from the Crystal Falls
district two, from Marquette County three, from the Baraboo
district, Wisconsin, two, and from Ontario one. Total, 24.
Two methods are in general use in the determination of iron.
The permanganate method, familiar to most of us, is used by
seventeen chemists; the other, the bichromate method, by but
seven.
* A paper read by W. A. Siebenthal before tlie Lake Superior Mining
Institute, Iron Mountain, Mich., meeting, October, igo5.
Metallurgical Azotes and Comments 551
Three Methods for Phosphorus
In the clctcrniination of phosphorus three general methods
are described with varying modifications in manipulation. One,
the Handy alkalimetric method, in which the phosphorus is
precipitated as yellow ammonium phosphomolybdate dissolved
in standard sodium hydrate and titrated with standard nitric
acid, is used by twenty of the chemists. The Emmerton method,
in which the yellow precipitate is dissolved in ammonium hy-
drate, reduced with zinc and sulphuric acid and titrated with
potassium permanganate, is used by three; and a modification
of the Wood method, described in Blair's " Chemical Analysis of
Iron," in which the phosphorus is determined gravimetrically by
weighing the yellow ammonium phosphomolybdate precipitate,
is used by one of the chemists. In some instances two methods
are reported by some of the chemists.
Two methods of more than the usual rapidity are described,
one by John McNamara of Ironwood, the other by F. A. Janson,
of Vulcan, Mich., both being modifications of the Handy method.
In the determination of silica two methods are described.
The sodium carbonate fusion method is used by seven; the
hydrofluoric acid method by four. Both are used by some of the
chemists.
Volhard's method for manganese, with various modifica-
tions, is used by fourteen of the chemists; Julien's method by
one, and a gravimetric process is described by one.
Of those reporting methods for the determination of calcium,
ten use a gravimetric method, precipitating the calcium as
calcium oxalate, igniting and weighing as calcium oxide (CaO).
One uses a volumetric process, titrating with potassium per-
manganate.
Magnesia is determined gravimetrically as magnesia
pyrophosphate by six, and alumina as aluminum phosphate by
nine of the chemists who report methods for such determinations.
Sulphur is determined as barium sulphate by those reporting
on the determination.
One chemist describes a method for the determination of
titanium.
Methods for moisture determination are described by four,
and for organic and volatile matter by three chemists.
d:)
The Iron and Steel Magazine
The Value of Comparisons
The interest and value of this collation of methods to the
chemists of the district, it seems to me, is to be obtained from the
differences in details and manipulation as well as in the general
differences of methods. Already I have found m3^self making
some changes in my own work, applying some of the details
given by some of the chemiists in their descriptions.
In conclusion, while some of the methods described may be
as rapid and reliable as any in general use, it seems to me that
there is a possibility for improvement and an excellent oppor-
tunity for research along the line of both shortening and simpli-
fying some of the methods given, especially in the determination
of phosphorus. A method for the direct oxidation and dissolving
of the phosphorus without the complete solution of the ore would
greatly simplify the process.
The two methods given for the determination of iron are
quite simple when compared with those described for other
substances ; yet each has certain objections — the permanganate
method, because of the inconstancy of strength of the solution;
the bichromate method, because it is slower and requires the use
of an external indicator.
An internal indicator would be a decided improvement in
the bichromate method. Keeping the permanganate solution
under some gas rather than in contact with the air might remedy
the objection to the permanganate method.
Phosphides and Carbides in Iron.* — At the opening meeting
of the session 1905-6, held in the Department of Applied Science
at Sheffield University on the 25th inst., Mr. J. E. Stead, in the
course of a lecture on '' The Behavior of Phosphides and Car-
bides when together in Iron," described the micro-structure and
mechanical properties of a series of steels containing 0.04 per
cent, 0.30 per cent and 0.50 per cent phosphorus and 0.30 per
cent carbon, showing that the most phosphorized portions were
concentrated at the junction of the primary crystals, and that
these junctions are always free from carbon in the annealed
material. A comparison of the results of testing by various
methods led to the interesting conclusion that ordinary tensile
* " The Iron and Coal Trades Review," November 3, 1905.
Metallurgical Notes and Comments 553
testing was not stifificient to indicate '' phosphorus brittleness,"
and tliat Professor Arnold's vibratory strain tests indicated an
increasing brittleness with each increment of phosphorus. Mr.
Stead described the method of shghtly bending small strips of
steel backwards and forwards till they broke, b\' which means
results closely corresponding to those of Professor Arnold were
obtained. The most remarkable results were those given by
testing the steels in a fiber stress machine of the Wohler type,
under a constant rotary stress of 25 tons. The resistances to
this true fatigue test for the three cases were in the ratio of i,
2.7 and 10. The steel with 0.5 per cent phosphorus stood ten
times as much fatigue as the steel with 0.04 per cent of that
element. \A'hy this was so was explained by the evidence given
by the tension tests, which showed that each o.i per cent phos-
phorus had raised the yield point and tenacity by about 2.5
tons per square inch. It was not to be supposed that he advo-
cated high phosphorus steels. The ingots he had used were only
6-inch cubes; had they been of the usual size employed in steel
works the results would have been quite different, due to ex-
cessive segregation of the phosphorus into local positions. The
lecturer went on to explain what occurred when steel solidified in
the ingot mold, and proved the existence of three distinct modes
of segregation : First, the microscopic in the body of the crvstal
itself; second, the minor segregations which were entangled at
the junctions of the crystallites; third, the major or axial segre-
gation well-known to all steel makers. The lecturer suggested
that if large ingots of fluid steel were revolved on their axes
during the cooling period, probably '' ghost " lines would not
appear in the forging, or if they did they would be very small,
the reason for this belief being that motion of the fluid steel
would tend to the production of small entanglements of the
segregated portions, whereas illustrations were given showing
that sometimes the crystals in very large ingots were as much as
7 inches in length.
The Harvey Steel Royalty Cases.* — By a judgment just
handed down by the Court of Claims, the Carnegie Steel Company
has been awarded $8,024.45 for royalties paid to the Harvey
Steel Company for the use of the so-called Harvey process for
* " The Iron Age," November 2.^, 1905.
^54 The Iron and Steel Magazine
face hardening armor. This process was employed by the
Carnegie Company under the terms of a contract with the
United States in which it was agreed that the government should
pay all royalties, but which it subsequently refused to do.
The Harvey process of hardening armor plate was employed
by the Ordnance Bureau of the Xavy Department for a number
of years on a rovalty basis. The government finally refused to
pay royalties and withheld a considerable sum from the paten-
tees on the ground that the patent was invalid, although it was
conceded that the process was efficient and was necessary to the
maniifacture of plates possessing the highest ballistic resistance.
The Harvev Steel Company, the owners of the process, then
began an action in the Court of Claims to recover 860,806.45
alleged to be due in the form of moneys withheld under con-
tracts with the Navy Department. The Court of Claims re-
cently decided this case adversely to the government, on the
ground that after having employed the process and having
received from the Harvey Steel Company all the information
necessary to produce the best armor plate known to the art, it
could not claim not to have received full consideration for the
payments promised under the contract. In conclusion, the
court laid down a proposition having a broad bearing upon
patent litigation in which the question of the validity of a
patent is involved in a claim for royalties, saying:
^' In a word, this is a case where a man without fraud or
misrepresentation entered into a contract; where he received
from the other party all that the contract promised him or that
he expected to receive ; where he kept his mouth closed when he
should have spoken and withheld the defense when he should
have interposed it ; where by his silence and his words he misled
the other contracting party and thereby deprived him of his
legal right to the adjudication of courts of competent jurisdiction,
which adjudication might be favorable to the other party and
cost irreparable loss and injury to himself. Such a man is not
entitled to set up in an action on the contract the defenses
which the defendant's executive officers have insisted on inter-
posing in this case. The court has not entered into an examina-
tion of the patent ; of the construction which should be given it ;
of the state of the art or of any of those questions which would
properly be s-ubjects of consideration if this were an action for
Metallurgical Notes and Comments 555
infringement." The eourt thereupon gave judgment for the
entire sum claimed.
In addition to this suit of the Harvey Steel Company
against the United States, the Carnegie Steel Company began an
action for $8,024.45 on account of royalties paid the Harvey
Steel Company for the use of its process in the mantifacture of
armor for the government. Before this case was taken under
consideration by the Court of Claims the United States Supreme
Court rendered a decision sustaining the court below in the case
of the Harvey Steel Company v. United States. The Court of
Claims therefore promptly gave judgment in favor of the Carnegie
Steel Company and will certify the decision to Congress for an
appropriation at the coming session.
Upon a supplemental petition filed by the Harvey Steel
Company the Court of Claims has given an additional judgment
against the United. States for $650,132.17, being the amount
of rovalties that have accrued since June, 1898, and up to
September 30, 1905, on the basis of the findings in the original
case of the Harvey vSteel Company v. United States above re-
ferred to.
Present Available Capacity of the Blast Furnaces of the
United States.* — The annual capacity of all the blast furnaces
in the United States which were active on June i, 1904, or which
were likely to be some day active, as published in our Directory
in August, 1904, amounted to 28,114,000 gross tons of pig iron.
Included in these figures were 8 completed furnaces, with an
annual capacity of 1,155,000 tons, which had not made pig iron
down to the date named but have since been blown in. Also
some furnaces which would never run again but whose fate could
not then be determined.
Since June i, 1904, 15 furnaces, with a total annual capacity
of 1,982,000 tons, have been completed and blown in, and 15
furnaces, with a total annual capacity of 461,000 tons, have
been abandoned or dismantled. In addition, 16 furnaces, with
a total annual capacity of 1,830,000 tons, were in course of
erection on November i, 1905. Full particulars concerning all
completed building and abandoned furnaces will be found in
* " Bulletin," American Iron and Steel Association, December i,
1905.
556 The Iron and Steel Magazine
accompanying" lists, which have been corrected to November i
or to later dates.
During the period between June i, 1904, and November 1,
1905, a number of furnaces which were classified as active on the
former date have been equipped with additional blowing ma-
chinery or have been rebuilt or reconstructed, thus increasing
their annual capacity, as Vv^e estimate it, at least 500,000 tons.
On the other hand, a number of furnaces which were included in
the active list on June i, 1904, have since been idle and are likely
to long remain idle, while others that have been idle since that
date will soon resume operations. We estimate the annual
capacity of these idle furnaces at approximately 1,500,000 tons.
vSummarizing the foregoing details we reach the following
results ;
Furnaces — Gross Tons Annual Capacity
Completed furnaces on June i, \(;o\ 28,114,000
Completed and blown in since June i, ic,o4 1,982,000
Total 30,096,000
Abandoned or dismantled since June i , 1904 461,000
Total 29,635,000
Rebuilt and enlarged since June i, 1904 500,000
Total 30.135.000
Furnaces idle since June i, 1904 1,500,000
Approximate annual capacity in November, 1905 28,635,000
The 16 furnaces which were in course of erection on Novem-
ber I, 1905, will have a total annual capacity of 1,830,000 tons.
Of these furnaces 3 stacks, with an annual capacity of 315,000
tons, will probably be ready for blast in January, 1906 ; 2 stacks,
with an annual capacity of 240,000 tons, will be ready in Feb-
ruary; 2 stacks, with an annual capacity of 215,000 tons, will be
ready in March; i stack, with an annual capacity of 145,000 tons,
will be ready in April ; 3 stacks, with an annual capacity of 450,000
tons, will be ready in June; and 5 stacks, with an annual capacity
of 465,000 tons, will be ready in the summer or fall of 1906.
As the production of pig iron in this country in October
was at the rate of almost 25,000,000 tons annually, it will be seen
that, if the capacity of the furnaces which have been idle since
June I, 1904, be omitted, over 87 per cent of the available fur-
nace capacity of the country on November i was then in operation.
Metallurgical Notes and Comments 557
Durini^ Xovenibcr several furnaces which have been idle
since June, 1904. have resumed work, and repairs are now being
made to a number of furnaces long idle which are expected to be
ready for blast in December, January and February. If to the
capacity of these furnaces we add the capacity of the building
furnaces which are to be ready for blast in January and Febru-
ary, 1906, it will l)e found that within the next ninety days
furnaces with an annual capacity of about t, 000, 000 tons will
probably be running.
The Crucible Steel Company of America.* — The fifth annual
report of the stockholders of the Crucible Steel Company of
America for the year ending August 31, 1905, was issued on
October 6. The report is very favorable. At the close of the
fiscal year, August 31, 1904, the total debt of the company was
$6,203,767.06; at the close of the fiscal year, August 31, 1905, it
was $3,650,189.87; showing a reduction of $2,553,577.19, in-
cluding $33,000 of bonded indebtedness. During the past year
the company has met all its current obligations and in addition
has anticipated payment of notes covered by the collateral trust
bonds to the amount of $808,000. This improvement in the
com.pany's financial condition renders it unnecessarv to issue
any part of the $7,000,000 bonds authorized at the last annual
meeting. There are now no liens of any kind against the plants
or properties of the company other than the collateral trust
bonds, amounting to $1,567,000, and the mortgages, amounting
to $171,932.56. The amount of these mortgages represents the
purchase price of several pieces of land at Clairton, now under
agreement of sale to the Clairton Steel Company.
Since the last annual statement the company has purchased
the outstanding stock (about 25 per cent) of the Canton Steel
Company and now owns all that company's property and assets.
A Large Steel Rail Contract.! — The Lackawanna Steel
Company has been awarded the record-breaking contract for
steel rails which are to be used in the construction of the Havana
Central Railway Company's extensive electric traction system
* " Bulletin," American Iron and Steel Association, November i, 1905.
■\ Ibid., November 15, 1905.
558 The Iron and Steel Magazine
in Cuba. The contract calls for the shipment of nearly 18,000
tons of rails, to be laid on almost 250 miles of track. This is by
far the most important contract ever let in this country for
trolley rails to be shipped outside the United States, although
there have been a number of larger orders placed in. the states
for foreign steam roads, notably for the Russian, Australian and
Argentina lines.
The Baldwin Locomotive Works.* — The Baldwin Loco-
motive Works built 225 locomotives in October last. This is the
largest monthly output in the company's histor}''. The total
output this year will break all records. To build these loco-
motives required the employment of 16,750 men in the shops in
Philadelphia alone. With the force at Burnham there are on
the pay rolls of the company more than 19,000 men. The
shops are kept going from midnight Sunday until midnight the
following Saturday. Most of the locomotives wxre large freight
engines. During the twenty-five working days in October nine
locomotives a da}^ were turned out.
The United States Steel Corporation.* — On October 31 the
United States Steel Corporation directors declared a regular
quarterly dividend of ij per cent on the preferred stock. The
unfilled orders on hand on September 30, 1905, aggregated
5,865,377 tons, against 4,829,655 tons on June 30, 1905, and
3,027,436 tons on September 30, 1904. The net earnings for the
September quarter of 1905 were $31,240,582, an increase of
$12,466,650 over the same quarter of 1904, when the earnings
were $18,773,932. In the same quarter of 1903 they were
$32,422,955.
Joint Meeting of the American Institute of Mining Engineers
and the Iron and Steel Institute. — The following is extracted
from a circular issued November 23, 1905, by the secretary of
the American Institute of Mining Engineers :
'' After much consultation and correspondence, the date
first suggested has been changed. It is now settled that the
meeting will begin in London, July 23, 1906, and will continue
(including all sessions, entertainments and excursions in Great
* " Bulletin," American Iron and Steel Association, November 15,
1905.
Metallurgical Notes a)id Comments 559
Britain) about two weeks. No attempt will be made through
this office to organize a special party for the ocean trip ; but if it
should hereafter appear that a sufficient number of members
desire to go and return in a body, the matter will be placed in
the hands of some leading tourist agency for the necessary
arrangements.
'' Comparatively little time will be available at this meeting
for the reading and discussion of papers — not more than two
or three sessions for the papers of both societies. Consequently,
the following niles and suggestions are issued, by authority of
the Council, for the acceptance and printing of papers, and their
presentation during the time allotted to this Institute. These
do not apply to ordinary papers, on subjects outside of iron and
steel. It has been agreed between the two societies that each
shall be free to publish in its own Transactions, in full, in part,
or in abstract, such papers, presented by the other, as it may
desire. Papers of this Institute, not relating to iron or steel, will
be ' read by title ' only at the joint sessions. But of such papers,
if received, accepted and published according to the conditions
stated below, pamphlet copies will be taken to London, for dis-
tribution to those interested in them, and to the technical
journals.
" Rules and Suggestions
'^ I. All papers presented by this Institute at the joint
sessions must have been printed beforehand. No such paper
will be read in full. The author, if present, will be allowed fifteen
minutes to indicate orally the important features of his paper,
and an equal time to close its discussion.
''2. Contribution to the discussion of any such paper may
be made in writing or orally; and the time allowed for their
reading or oral presentation will be decided between the secretary
and the author, with the understanding that, as far as possible,
no one desiring to participate in the discussion shall be excluded
by reason of the allowance of more than ten minutes to any
other; that all remarks made in oral discussion (modified or
amplified as the author may desire) shall be, immediately there-
after, furnished to the secretary in writing; and that acceptable
discussion, whether presented at the meeting or not, will be sub-
sequently published by the Institute.
'' 3. In order to secure valuable discussion, the papers
560 The Iron and Steel Magazine
presented bv this Institute should be pubUshed in preUminary
form, and distributed to members of both societies, before the
date of the meeting. Consequently, all papers concerning iron
or steel, intended for such presentation, must be in the hands of
the secretary before March 31, 1906, so that they may be ex-
amined, accepted, edited, published and distributed by June i,
1906. These are the latest permissible dates. Those who can
furnish manuscripts earlier (especially if they reside at a con-
vsiderable distance from New York City, or if their papers are
accompanied with illustrations), will greatly facilitate the work
of the secretary and the printer by doing so.
'^4. Members proposing to offer papers of this class are
earnestly requested to give immediate notice (stating the nature,
length, etc., of the proposed papers) to the secretary, who will
take pleasure in facilitating, by suggestion and advice, the work
of their preparation. In view of the now acknowledged leader-
ship of the United States' in the production of iron and steel,
papers descriptive and critical of modern American practice are
deemed specially appropriate and desirable.
''5. All the papers relating to iron or steel, received and
accepted before March 31, 1906, will be printed as soon as pos-
sible, and distributed at or before the joint meeting, but prece-
dence wall be given in order of printing to those selected as most
important and appropriate, and hkely to elicit the widest and
most profitable discussion. These will be sent in advance to the
members of both societies interested in their subjects, and quali-
fied to criticise them intelligently.
''6. It is impossible to say beforehand exactly how many
accepted papers Vvdll be presented for actual discussion at the
joint meeting. Probably the thorough discussion of three, or
perhaps four, will exhaust the available time. But the rest will
be presented in print, with brief reference to their nature, and in-
vitation of appropriate discussion by mail; and will be included,
together with such discussion, in the Transactions of this Insti-
tute, and also placed at the disposal of the Iron and Steel Institute,
for full or partial publication in the journal of that society.
'' This occasion offers, therefore, an unusual opportunity,
not only for contributions of international value to the arts and
sciences connected with iron and steel, but also for the winning
of international recognition for such services."
REVIEW OF THE IRON AND STEEL MARKET
There has been a general advance of from $r.oo to $1.50 in
pig iron in aU markets since our last report, while crude steel and
finished steel products are substantially unchanged in price.
There are expectations that pig iron will score further advances
before the end of the year, reaching possiblv $20 at furnace.
Several finished steel products would have advanced before this
had not the large producers steadfastly set their faces against
such a movement.
Buying of finished steel products has in general exceeded
production during the month, and mills are sold farther ahead
than they were two months ago. Production, however, has been
extremely heavy. If the October rate of pig-iron production is
maintained to the end of the year, production will have amounted
to about 22,900,000 gross tons for the whole year, or about
4.900,000 tons in excess of the best previous record. All this
pig iron is going into immediate consumption.
Since the Pennsylvania railroad system placed orders for
21,000 steel freight cars, as noted in last report, the New York
Central has bought 25,000 steel cars, and a number of smaller
orders have been received, so that the steel car plants of the
country are sold up through the first nine months of next year.
The plate and shape orders against this work have been placed.
Pig Iron. — Buying of pig iron has been fairly large in all
markets, although there has not been the rush which character-
ized the market in October. As a result of that movement, how-
ever, furnaces were so well sold up that on the light sales during
November the market has advanced even more rapidly than it did
in October. Southern pig iron has advanced about $1.50,
northern foundry iron from $1.00 to $1.25 and Bessemer about
$1.50. There is little iron available either north or south for
delivery before the second quarter of next year. Prices are very
firm as follows: F.o.b. valley furnace: Bessemer, $17.50; basic,
$17.25; No. 2 foundry, $17.00 to $17.50; forge, $16.00 to $16.25.
561
562 The Iron and Steel Magazine
Delivered Pittsburg: Bessemer, $18.35; basic, $18.10; No. 2
foundr^^ $17.85 to $18.35; foi*ge, $16.85 ^o $i7-io- F-O-b.
Birmingham: No. 2 foundry, $14.50; gray forge, $13.25. De-
livered Philadelphia: No. 2 X foundry, $18.00 to $18.50; stand-
ard grav forge, $16.25 to $16.50. Delivered Chicago: northern
No. 2 foundry, $19.00 to $19.50; malleable Bessemer, $19.00 to
$19.50; Lake Superior charcoal, $19.50 to $20.00. Freight:
Birmingham to Pittsburg, $4.35; to Cincinnati, $2.75; to
Chicago, $3.65; to Philadelphia by water, $3.50; by all-rail,
$4.00.
Ferro-Manganese. — Largely on account of the troubles in
the Caucasus, the supply of mianganese ores has been reduced at
a time when the demand is unusually heavy. Deliveries of ferro-
manganese on contracts have been poor, and a number of con-
sumers have been forced into the market at a time when there
was scarcely any prompt metal to be had, the result being an
unprecedentedly rapid advance. About the first of November
the fairly large lots available for early delivery w^ere sold, at
about $65, delivered, for 80 per cent metal. Since then on sales
of small lots the market has advanced rapidly, $75 being touched
in the first fortnight of the month and $90 in the third week, and
the market cannot now be definitely quoted. Late deliveries
would not bring such high figures, but there is no interest in such
deliveries at present, as contracts are not being filled as they
should be and it is uncertain when there will be relief from the
scarcity.
Steel. — The market has been almost nominal through the
month, with no particular change in prices. A number of
exchange deals have been made, but no large tonnages have sold
in the open market. We continue to quote prices as follows:
F.o.b. Pittsburg: Bessemer billets, $26; open-hearth, $27 to $28;
forging billets, $30; sheet bars, long lengths, $27; wire rods,
$32; chain rods, $33.
Rails. — There has been further booking of large rail con-
tracts, and the western and southern mills are sold for prac-
tically the entire year 1906. The price for standard sections
remains at $28 at mill. Light rails are scarce for any early
delivery, and command a premium in some cases above the
regular prices for later delivery, based upon $26 at mill for
sections 25 to 45 pounds per yard.
Review of the Iron and Steel Market 563
SJiapes. — A large tonnage continues to be booked. The
large structural mills are sold almost through the first half of
next vear. with actual specifications in hand on most of the ton-
nage. Production will be increased in December and January
by the completion of two new mills, one at Clairton by the
Carnegie Steel Company and one at the South Chicago works of
the Illinois Steel Company. These mills will have an output
roughly estimated at about 10,000 tons each per month. Prices
are unchanged for regular mill delivery, based on 1.70 cents for
beams and channels, 15-inch and under, angles 2 x 3 to 6 x 6
inclusive and zees.
Plates. — The plate mills have large tonnages sold against
steel car work through the first three qtiarters of next year, and
are sold for their entire output until some time in the first quar-
ter. The regular mill price remains at 1.60 cents for tank qual-
ity, quarter-inch and heavier.
Merchant Bars. — Iron bars have advanced in all markets,
bv about $2.00 a ton in the east and $1.00 a ton in the west.
The market stands at t.8o cents f.o.b. Youngstown and 1.85
cents, delivered, at Pittsburg and Chicago.
Sheets. — The American Sheet and Tin Plate Company
advanced its prices on all sheet products 10 cents per hundred
pounds, effective November 20. Many of the independent mills
had already started quoting higher prices. An advance of 10
cents a box was made in tin plates. Practically all large buyers
had covered before these advances, but they are holding as
regards such business as comes up. We now quote on No. 28
gauge sheets, f.o.b. Pittsburg, 2.30 cents minimum on black and
3.35 cents minimium on galvanized, with single carloads some-
times bringing 5 cents a hundred additional. Tin plates are now
$3.35 per box, net, for 100-pound cokes, f.o.b. Pittsburg.
Scrap. — The market has advanced rather sharply on light
sales, the dealers absorbing scrap willingly and holding it for
winter, when much higher prices are expected to prevail. We
quote the market approximately as follows: Heavy melting
scrap, $18.00; cast borings, $11.00; sheet scrap, $15.50; No.
I wrought scrap, $20.00; No. i cast scrap, $15.50, all dehvered,
Pittsburg district.
Coke. — There has been no further advance in Connellsville
coke on contracts for 1906 since the high prices which had been
564 The Iron and Steel Magazine
reached at the time of our last report, while Connellsville furnace
coke for prompt shipmient has declined a trifle on account of two
or three furnaces being temporarily out of blast. We quote for
strictly Connellsville coke, net ton at ovens, $3.00 on furnace
coke on contract and $2.90 to $3.00 on prompt coke, and $3.50
on foundry coke on contract and $3.75 to $4.00 for prompt
shipment.
STATISTICS
German Pig-iron Production. — The returns of the Associa-
tion of Iron and Steel Producers show that the German pro-
duction of pig iron in September and in the first nine months of
the year has been as follows, compared, with last 3^ear, the
figures referring to metric tons of 2,204.6 pounds :
Sex^tember First Nine Months
1904 1905 1904 1905
Foundry iron ..... 163,302 168,841 1.359.345 i. 379. 000
Bessemer iron .... . 23,175 34,634 310. 79° 315.614
Basic iron 523,042 618,472 4,777,728 5,170,527
Spiegeleisen 53.4i2 65,185 459.534 501,005
Forge iron 70,677 66,648 623,572 597,450
Totals 833,608 953.780 7,530,969 7.963.596
Belgian Pig-iron Production. — Pig-iron production in
Belgium in September showed an increase of 4,622 tons over the
production in September of last year, making the gain for the
nine-month period 23,835 tons. The production by grades has
been as follows, in metric tons of 2,204.6 pounds :
First Nine Months
1904 1905
Bessemer and basic 716,151 764,746
Forge pig 183,037 161,153
Foundry pig. 79,150 76,274
Totals 978,338 1,002,173
World's Pig-iron Production. — Official statistics have
become available of pig-iron production in Sweden in 1904,
showing the total to have been 528,525 metric or 520,173 gross
tons. An estimate is made of Italian production in 1904 at
120,000 tons. Statistics of 1904 pig-iron production have been
available for some time for the United vStates, Germany, United
Kingdom, France, Russia, Belgium and Canada, and such pro-
duction added to the production in Sweden and Italy makes a
total of 42,969,810 gross tons, with Austria-Hungary and Spain
565
566 The Iron and Steel Magazine
vet to be heard from, and an estimate to be made for various
countries which do not make statistical returns. In 1903 such
countries were credited with 1,918,462 tons, so that should there
be no change in this direction the world's production of pig iron
in 1904 would stand at 44,888,272 gross tons, against 45,894,713
gross tons in 1903, 43,324,068 gross tons in 1902, 40,100,000
gross tons in 1901 and 40,087,616 gross tons in 1900. The
decline in the world's pig-iron production in 1904 was due to the
United States and the United Kingdom, the former showing a
decrease of 1,512,219 gross tons and the latter a decrease of
248,546 gross tons.
Russian Iron and Steel Statistics.* — ■ Through the courtesy
of Mr. Adolphe Wolski, of the Statistical Department of the Min-
istry of Finance of the Russian Government, we are enabled to
give below the official statistics of the production of iron ore,
coal, pig iron, steel ingots and castings, and steel rails in Russia
and Finland for the last fifteen years.
Iron Ore. — The maximum production of iron ore in the
Russian Empire was reached in 1900, when 6,112,090 metric
tons were mined. The next highest years were 1899, when the
production amounted to 5,890,900 tons, and 1904, w^hen it
amounted to 5,272,300 tons. In the following table the pro-
duction of iron ore is given, in metric tons, from 1890 to 1904:
Years
Metric Tons
Years
Metric Tons
Years
Metric Tons
1890
1,795.663
1895
2,986,715
1900
6,112,090
1891
1,958,452
1896
3,321,786
1901
4,723,983
1892
2,044,106
1897
4,102,536
1902
3,987,303
1893
2,210,305
1898
4,536,217
1903
4,218,600
1894
2,524,610
1899
5,890,900
1904
5,272,300
Coal. — The production of coal in the Russian Empire was
as follows from 1890 to 1904. The maximum production was
reached in 1904.
Years
Metric Tons
Years
Metric Tons
Years
Metric Tons
1890
6,085,080
1895
9,098,800
1900
16,135,600
1891
6,223,450
1896
9,384,900
1 90 1
16,507,240
1892
6,946,200
1897
11,210,524
1902
16,431,440
1893
7,610,600
1898
12,333,500
1903
17,818,000
1894
8,762,600
1899
14,311,200
1904
19,318,370
* " Bulletin," American Iron and Steel Association, September 15, 1905.
Statist it's 567
Pig Iron. — xVlthough the production of pig iron in the
Russian Empire in 1904 was the largest in its history the in-
dustry has made bvit Uttle progress statistically since 1899, in
which year the production was only 250,943 tons less than in
1904. In the following table the production of pig iron is given
since 1890. Russia is a large importer of pig iron, her production
in some years being only slightly in excess of her annual produc-
tion of steel ingots and castings.
Years
Metric Tons
Years
Metric Tons
Years
Metric Tons
1890
927,214
1895
1-453,529
1900
2,936,191
1891
1,005,570
1896
1,607,490
1901
2,869,306
1892
1,072,651
1897
1,881,670
1902
2,597,435
1893
1,147,391
1898
2,243,081
1903
2,486,610
1894
1,348,615
1899
2,727,382
1904
2,978,325
Steel Ingots and Castings. — In the following table will be
found the production of all kinds of steel ingots and castings in
Russia from 1890 to 1904. Over two thirds of the steel now
made annually in the Russian Empire is produced by the open-
hearth process. The production of steel castings by the Besse-
mer and open-hearth processes amounts to from 35,000 tons to
40,000 tons annually. About 2,000 tons of steel ingots and
castings are also annually produced by the crucible process. In
addition, from 5,000 tons to 6,000 tons of steel are annually made
by various minor processes. The growth of the Russian steel
industry from 1890 to 1904 has been remarkably strong and
steadv.
Years
Metric Tons
Years
Metric Tons
Years'
Metric Tons
1890
378,422
1895
880,451
1900
2,217,752
1891
440,212
1896
1,023,118
1901
2,230,000
1892
516,315
1897
1,225,526
1902
2,183,400
1893
633,120
1898
1,621,801
1903
2,410,938
1894
726,017
1899
1,903,000
1904
2,81 1,948
Steel Rails. — The production of steel rails in Russia has
been as follows, in metric tons, from 1890 to 1904. The maxi-
mum production was reached in 1900, when 496,475 tons were
rolled.
Years
Metric Tons
Years
Metric Tons
Years
Metric Tons
1890
166,156
1895
302,428
1900
496,475
1891
166,503
1896
398,848
1901
481,918
1892
193,338
1897
444,062
1902
382,152
1893
230,954
1898
468,787
1903
332,367
1894
250,190
1899
464,377
1904
401,541
5 68 The Iron and Steel Magazine
Iron and Steel in Sweden.* — According to the report of
Director-General Richard Akerman on the Swedish iron industry,
the production of iron ore, which was 3,677,841 metric tons in
1903, was 4,084,647 tons in 1904, showing an increase of 406,806
tons, or 1 1. 1 per cent. The production of coal in 1904 was 320,-
984 metric tons.
The output of pig iron was 506,825 metric tons in 1903, and
528,525 tons in 1904, an increase of 21,700 tons, or 4.3 per cent.
There were 133 furnaces in blast in 1904; the average active
period was 263 days, giving an average yield per furnace of 3,974
tons for the year, or 15.1 tons per day. The Swedish furnaces all
use charcoal as fuel. The qtiantity of blooms made from pig
iron in charcoal hearths — an industry now almost peculiar to
Sweden — was 189,246 tons in 1904.
The production of steel ingots and direct castings was as
follows, in metric tons :
1903 1904 Changes
Bessemer. 84,229 78,577 D- 5.652
Open-hearth. 232,878 252,832 I. 19.954
Total 317.107 331.409 I. 14,302
Last year there were reported, in addition, 1,162 tons of
crucible steel ingots and 951 tons of bhster steel, bringing the
total make of steel of all kinds up to 333,522 metric tons. The
ratio of steel to pig iron last year was 0.63, a high figure.
The production of finished iron and steel in various forms is
reported as follows, in metric tons:
Bars • 181,775
Nail-plates and wire-rods 102,976
Other shapes 9,020
Plates 16,331
Tube-blocks and hollow-blooms 23,594
Total 333.696
Adding the charcoal blooms made direct from pig iron gives
a total of 522,930 tons of finished or semi-finished products. In
this total, however, there is some dupHcation, as a part of the
charcoal blooms is converted into bars and wire-rods, though a
quantity is also sold and exported in the form of blooms.
* " Engineering and Minin.2: Journal," October 7. 1905.
Statistics 569
The high quahty of Swedish iron makes a strong export
deiiiaiul, not onlv for these bloonis, but also for bars, wire-rods
and other forms of iron and steeL
British Half Yearly Steel Statistics.* — The total output of
Bessemer steel ingots in Great Britain in the first half of 1905, as
ascertained bv the British Iron Trade Association, was 1,019,887
tons, which compares with 865,683 tons in the first half of 1904
and with 911,670 tons in the first half of 1903.
The production of Bessemer steel rails in the first half of
TQ05 was 540,314 tons, against 523,771 tons in the first half of
1Q04. It will be observed that the output of steel rails has not
increased to the same extent as the output of ingots, and that the
quantity of steel rails produced is not much over one half that
of the Bessemer ingots produced in the first half of 1905.
The output of open-hearth steel ingots in the United King-
dom in the first half of 1905 has been ascertained by the British
Iron Trade Association to have been 1,980,095 tons, which is an
increase of 309,966 tons on the production in the corresponding
period of the previous year, or at the rate of nearly 620,000 tons
of increase for the whole twelve months.
The output of open-hearth steel ingots in the first half of
1905 is considerably in excess of that of any previous half year,
the nearest approach to it having been an output of 1,710,602
tons in the first half of 1902.
Production of Natural Gas in 1904.* — F. H. Oliphant has
prepared a paper for the United States Geological Survey on the
production of natural gas in this country in 1904. This pro-
duction amounted approximately to 256,645,000,000 cubic feet,
or 6,159,480 tons of 2,000 pounds. The value of this production
was $38,496,760, w^hich was an increase of $2,688,900 over the
value of the production in 1903. The production of Penn-
sylvania alone was valued at $18,139,914, or 47 per cent of the
entire value. Four States, Pennsylvania, West Virginia, In-
diana and Ohio, produced 93.3 per cent of the entire value of
natural gas produced in 1904. The total production in 1904
was greater than that of any previous year. Natural gas was
produced in 1904 in twenty states and territories.
* " Bulletin," American Iron and Steel Association, November 15,
1905-
570 ^^^^ Iron and Steel Magazine
The value of natural gas produced and consumed in the
United States in the last eight years has increased nearly three-
fold. It is remarkable that Pennsylvania, which was the first
State to use natural gas on a large scale, has regularly- maintained
a yearly increase for the last eight years, the product for 1904
being about three times the quantity marketed in 1897. West
Virginia and Ohio have also shown a remarkable increase in the
production of natural gas since 1897. Indiana has for the last
two years shown a heavy decline in production. Alabama has
for the first time entered the list of natural-gas producing States.
Some natural gas is produced in Canada and consumed in the
United States but is not included in the above statistics.
The large increase in the value of the production of natural
gas in Pennsylvania in 1904 is remarkable when it is remembered
that Pennsylvania is the oldest State producing natural gas in
any large quantity. The increase has been derived from the
deeply buried sands in Greene and Washington counties in the
southwestern portion of the State, and from the counties of
Armstrong and Clarion, where deeper producing sands have been
developed in 1902 and 1903. A few small pools were secured
in Potter County.
Some very large natural-gas wells were developed in Kansas
in 1904, whose discovery was in part due to the large amount of
drilling that was done in search of petroleum, and which, from
all appearances, must soon put Kansas among the great natural-
gas producing States.
Natural gas is produced in California, but the total pro-
duction is comparatively small.
RECENT PUBLICATIONS
Sir Henry Bessemer, an Autobiography. 380 9 X ii-in.
pages; illustrated. Offices of " Engineering." London. 1905.
Price, $4.00. — We publish in full in this issue the chapter of this
book entitled, " The Genesis of the Bessemer Process," which
we consider a masterful description both as to clearness and
technical accuracy of the evolution of this epoch-making inven-
tion. It will suggest the pleasure and instruction to be derived
from the reading of Sir Henry Bessemer' s autobiography. Few
inventions, if any, have had a more momentous influence upon
our material civilization, and few inventors have shown greater
ingenuity, clearness of mind and logical sequence in bringing
to a triumphant climax the working of a sound conception.
Sir Henry Bessemer's biography is refreshing and inspiring
reading, relating as it does an eminently useful and successful
life, free from those melancholy chapters which form too often a
conspicuous part of an inventor's life. The book is divided
into twenty-one chapters bearing the following titles, the con-
cluding chapter having been written by the publishers with the
assistance of Sir Henry Bessemer's son : Early Days ; The Reward
of Invention; Compressing Plumbago Dust; Casting Type,
Type-Composing Machine, etc.; Utrecht Velvet; The Manufac-
ture of Bronze Powder; Improvements in Sugar Manufacture;
A Holiday in Germany; Improvements in Glass Manufacture;
The Exhibition of 185 1; Early Gunnery Experiments; The
Genesis of the Bessemer Process; The Bessemer Process; Besse-
mer Steel and Colonel Eardlev Wilmot; The Bessemer Process
and the War Office; Bessemer Steel; The Armstrong Contro-
versy; Bessemer Steel Guns; Cast Steel for Ship-building; Man-
ganese in Steel-Making; Ebbw Vale; The Bessemer Saloon
Steamship; Conclusion. The publishers are to be congratu-
lated in bringing forward so interesting a book in so attractive
a form, both the typography and engraving being a highly
creditable example of the best work of printer's and engraver's
arts. The book should appeal not only to metallurgists and
571
572 The Iron and Steel Magazine
engineers, but also to a very-large class of thoughtful and appre-
ciative readers.
Contribution a V Etude de la Fragilite dans les Fers et les Aciers
(Contribution to the Study of Brittleness in Iron and Steel).
482 9 X ii-in. pages; illustrated. Paper covers. Societe
d' Encouragement pour I'lndustrie Nationale. Paris. 1904.
Price, 20 francs, —This important volume is chiefly made up
of reprints from the bulletin of the " Societe d' Encouragement
pour r Industrie Nationale," dealing with the brittleness of iron
and steel and the testing m.ethods best adapted to its detection.
It is stated in a short preface that during the last fifteen years
numerous investigations have been conducted, especially in
France, into the brittleness of steel. They appeared in various
publications and some of them are to-day difficult to procure.
The " Societe d' Encouragement " thought that a reprint of these
articles in book form would be of much value to metallurgists
and constructing engineers. The society was assisted in its
task by six large railway companies, to whom her thanks are
due for their generous support. The articles reprinted here are
by Messrs. Ast, Aucher, Barba, Brinnell, Brustlein, Charpy,
Andre Le Ch atelier, Henry Le Chatelier, Considere, Fain, de
Freminville, Fremont, Godron, Guillery, Huillier, Leblant,
Mesnager, Osmond, Ridsdale, Vanderheym and Wahlberg.
Under the title of " Brittleness of Steel," and as an introduction to
the book, Prof. Henry Le Chatelier gives a very able and instrtic-
tive resume of the most important results of these investigations.
The Societe d' Encouragement is certainly entitled to much credit
for her enterprise and her leadership in promoting the scientific
studv and scientific treatment of the metal which forms the very
foundation of our material civilization.
Technical Methods of Ore Analysis, by Albert H. Low. 273
6 X 9-in. pages; illustrated. John Wiley & Sons, New York.
1905. Price, $3.00. — The following extracts from the author's
preface will give an accurate idea of the scope and character of
the book:
" This book is primarily intended as an aid to the technical
chemist, but it is hoped it may also prove useful to the student
desiring to become acquainted with technical methods.
Recent Puhlications 573
"It is a common experience with technical chemists to
receive a sample of material with instructions to return the per-
centage of some constitutent whose technical determination is
more or less unfamiliar under the given conditions. In such a
case the chemist has recourse to his books and too frequently
is quite unable to find a method that is exactly adapted to the
material in hand, or that begins at the beginning and tells him
just how to proceed. He is thus left to work out his own sal-
vation, possibly at the expense of much valuaVjle time.
" In this book an attempt has been made to supply the want
thus indicated by describing methods that are adapted to the
cases most likely to be met in practice, although it is sometimes
practically impossible to devise a short technical method that
will meet every probable case.
" It has been my aim to make the descriptions so minute
and complete that if the operator will follow them exactly he
can scarcely fail to obtain satisfactory results.
" Some of the methods in the following collection have been
devised by myself, mainly on the basis of previously well-known
facts ; some are compilations of the work of others and some are
modifications of existing methods. I have endeavored to give
proper credit in all cases."
The methods appear to be very methodically and clearly
described and it certainly seems as if Mr. Low's book should be
found of much assistance by analytical chemists. The typog-
raphy of the book as well as the paper and binding are excellent.
Cours d' Exploit ation des Mines, third edition, by Haton
de la Goupilliere, with revisions and additions by Jean Bes de
Berc. Volume I. 1,0026^ X lo-in. pages; illustrated. Paper
covers. Vve. Dunod. Paris. 1905. Complete in three vol-
umes. Price, 90 francs. — This is the third and considerably
enlarged edition of a book of universal reputation. There are
probably very few living mining engineers who have not at least
consulted this classical work during their student days, and a
large number have used it as a textbook. In this new edition
the subject has been brought up to date by Jean Bes de Berc,
himself a distinguished engineer. It is to be complete in three
volumes. The first volume which we have before us deals with
the discovery of Mineral Deposits, Breaking Ground and Drilling,
574 ^^^ Iron and Steel Magazine
Shafts and Galleries. The second volume will deal with various
methods of mine exploitations and the third volume with Drain-
age, Ventilation and Ore Dressing. The publishers deserve
the gratitude of the mining professions for bringing forward this
modernized edition of so important a book.
Transactions of the Institution of Mining and Metallurgy.
Volume XIII (i 903-1 904). Edited by Arthur C. Claudct and
C. McDermid. 568 6 X 8^-in. pages; illustrated. Paper cov-
ers. E. and F. N. Spon. London. — This volume contains
the minutes of the meetings held during the thirteenth session
(i 903-1 904) of the Institution of Mining and Metallurgy. Among
the many valuable papers read before the Institution w^e note
the following as most likely to be of interest to our readers : The
Equipment of Laboratories for Advanced Teaching and Re-
search in the Mineral IndUvStries, by Henry C. Jenkins; Iron-
Ore Mining in vScandinavia, by W. Fischer Wilkinson; The
Microscope as an Aid in Copper Refining, by H. Nestor Schur-
mann ; A Graphic Method for the Computation of Blast-Furnace
Charges, by C. O. Banister.
Smithsonian Institution, annual report of the Board of
Regents for the year ending June 30, 1904. 804 6 X 9-in. pages;
illustrated. Government Printing Office. Washington, D. C.
1905. — The present issue of this valuable yearly publication
contains the usual number of reprints from important scientific
papers and fine illustrations covering a w4de range of subjects.
The Universal Directory of Railway Officials, igoj. Com-
piled under the direction of S. Richardson Blundstone, editor
of the "Railway Engineer." 667 5.7 X 8J-in. pages. — -The
Directory Publishing Company. London. Price, 10 shillings.
— This is the eleventh year of publication of this directory, the
value of which will readily be appreciated when it is considered
that it contains a list of the principal officers oi every railroad in
the world, with a brief description of the equipment, gauge and
mileage of each road. The trami lines of Great Britain are also
included. The contents are compiled entirely from official
sources.
PATENTS
RELATING TO THE METALLURGY OF IRON AND STEEL
UNITED STATES
801,136. Ingot Stripper. — Dwight B. Cheever, Chicago, 111.,
assignor to Whiting Foundry Equipment Company, Harvey, 111., a
cor])oration of Illinois.
801.143. Process of Preparing Dust Ores for Blast Furnaces.
— Heinrich E. Eich, Giessen, Germany, assignor to the firm of Fellner &
Ziegler, Frankfort-on-the-Main, Bockenheim, Germany.
801.144. Preparing Dust Ores for Blast Furnaces. — Johann
C. Fellner, Frankfort-on-the-Main, Germany, assignor to the firm of
Fellner & Ziegler, Frankfort-on-the-Main, Bockenheim, Germany.
801,229. Method of Casting Metals. — James Eaton-Shore,
Rugby, England, assignor to Willans & Robinson, Ltd., Rugby, England.
801,274. Process of Converting Steel-Scrap into Iron. —
Edward M. Schulz, Sergeants Hall, and George W. Helmlinger, Pittsburg,
Pa.
801,339. Compound for Hardening or Case-Hardening Iron or
Steel. — Gustav Reininger, Berlin, Germany, assignor to the firm of
Cyanid-Gesellschaft mit Beschraenkter Haftung, Berlin, Germany.
801,347. Water-Cooled Furnace Valve. — Francis H. Treat,
Pittsburg, Pa.
801,453. Molten-Metal Conductor for Cupolas, Etc. — Carmi
L. Glover, Newcastle, Pa.
801,500. Apparatus for Making Steel. — Frank E. Young,
Canton, Ohio.
801,646. Apparatus for Preparing Ingot-Molds for the Cast-
ing Operation. — Hugo Carlsson, Stockholm, Sweden, assignor of one
half to James H. Le Fevre, Buffalo, N. Y.
801,656. Process for the Deoxidation of Ingot-Iron, Ingot-
Steel, Etc. — Richard EichhofT, Essen-Riittenscheid, Germany, assignor
by mesne assignments to Elektrostahl Ges. M. Beschr. Haft., Remscheid-
Hasten, Germany.
801,842. Compound for Coating Iron and Steel. — Nicholas A.
Bibikov, New York, N. Y., assignor of one fourth to Henry Connett,
New York, N. Y.
801,947. Magnetic Separator. — John P. Wetherill, South
Bethlehem, Pa., and Henry A. J. Wilkens, New York, N, Y., assignors by
mesne assignments to Wetherill Separating Company, a corporation of
New Jersey.
802,030. Bell Operating and Controlling Mechanism for
Blast Furnaces. — John P. Coleman, Edgewood, Pa., assignor to the
Union Switch & Signal Co., Swissvale, Pa., a corporation of Pennsylvania.
802,131. Metal-Handling Device. — Henry Aiken, Pittsburg, Pa.
575
576 The Iron and Steel Magazine
802,151. Tilting Furnace.- — Francis H. Treat, Pittsburg, Pa.
802,170. Magnetic Ore-Separator. — Richard R. Moffat, Brook-
lyn, N. Y., assignor to Imperial Ore Separator Company, a corporation
of New York.
802,176. Blast Furnace. — Samuel B. Sheldon and Alexander K.
Hamilton, Buffalo, N. Y.
802,931. Gas-Producer. — Jerome R. George, Worcester, Mass.,
assignor to Morgan Construction Compan}^, Worcester, Mass., a corpora-
tion of Massachusetts.
803,337. Metallurgical Furnace. — Harry H. Goodsell, Leech-
burg, Pa.
So3'597- Device for Operating Furnace Doors, Gates, Etc. —
Joseph S. Hood, Stahlstown, Pa.
803.673. Apparatus for Rolling Sheet and Tin Plate. — Percy
E. Donner, Columbus, Ind.
803.674. Apparatus for Rolling Sheet and Tin Plate. — -Percy
E. Donner, Columbus, Ind.
803,745. Apparatus for Rolling Sheet and Tin Plate. — Percy
E. Donner, Columbus, Ind.
803,763. Bearing for the Rolls of Rolling Mills. — John T.
Horner, Newcastle, Pa.
803,886. Treatment of Iron Ores, Etc. — Carleton Ellis, New
York, N. Y.
804,023. Coupling Box for Rolls. — William C. Millhizer and
Frederick McQuiston, Pittsburg, Pa.
804,080. Apparatus for Making Steel and Other Metals. —
Edwin C. Wills, Trenton, N. J.
804,193. Process for Manufacturing Briquettes. — Gottfried
Hoepfner, Bleckendorf, Germany, assignor to Willy von Lewinski, Breslau,
Germany.
GREAT BRITAIN
7,599 of 1905. Refractory Lining for Furnaces. — J. Bach,
Riga, Russia. Making refractory linings for furnaces by mixing 20 parts
of pure hydrate of alumina with 80 parts of ordinary fire-clay and applying
the mixture, while moist, to the thickness of an inch and then burning.
9,774 of 1905. Blast-Furnace Blowing. — J. W. Dougherty,
Steelton, Pa. A method of blowing blast furnaces without the liability
of explosions.
9,199 of 1905. Recovery of Iron from Slag. — O. Thiel, Land-
stuhl, Germany. An improved furnace and process for recovering iron
from slags containing much oxide of iron.
28,295 of 1904. Blast-Furnace Gas Scrubber. — J. E. Thorny-
croft, London. Improved water sprays for removing dust and tarry
matters from blast-furnace and other gases.
11,875 of 1905- Blast-Furnace Water Jacket. — C. W. Hawkes
and F. Klepetko, Great Falls, Mont. Improved form of water jacket
for blast furnaces.
^
TN
1
18
v.lO
The Iron and steel magazine
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
ENGIN STORAGE
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11